Collaborative Control Method for Civil Construction Machinery Groups in Switchgear Stations

By using distributed interlocking judgment and magnetic induction backup verification methods, the problem of frequent mechanical collisions at the switchyard civil construction site was solved, achieving safe interlocking and construction continuity when communication is unreliable, and avoiding secondary accidents.

CN123113902APending Publication Date: 2026-11-17STATE GRID SHANDONG ELECTRIC POWER CO CONSTR CO +1
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202611610326.5
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-10-19
Publication Date
2026-11-17

AI Technical Summary

Technical Problem

Frequent mechanical collisions occur at the switchyard civil construction site. Existing anti-collision interlocks rely on unreliable wireless communication, which is prone to loss or misjudgment when communication fails. Furthermore, a complete shutdown can easily lead to secondary accidents, and it cannot take into account both construction continuity and differences in mechanical kinematics.

Method used

A method combining distributed interlock determination and magnetic induction backup verification is adopted. Each machine locally acquires and broadcasts status information, determines collision risk by relative proximity speed and dynamic safety distance, and performs backup verification using passive magnetic induction markers when communication fails, and outputs a shaft locking request.

Benefits of technology

It can maintain interlocking function even when wireless communication is unreliable or fails in a local area, avoid mechanical collisions and maintain construction continuity, prevent secondary accidents, accurately estimate dynamic safety distances, and avoid misjudgment.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN123113902A_ABST
    Figure CN123113902A_ABST
Patent Text Reader

Abstract

This invention relates to the field of construction machinery control, and more particularly to a collaborative control method for a group of civil construction machinery in a switchyard. The method includes: a distributed interlocking determination step, where each machine periodically broadcasts and maintains its proximity status; the relative approach speed is calculated from the position difference between two adjacent frames; a dynamic safety distance is calculated by combining the total response delay and the positioning error band; if the distance is not greater than the sum of the envelopes of both machines and the dynamic safety distance, a lock axis request is output; a magnetic induction backup verification step, where passive magnetic induction markers are pre-embedded; upon passing a marker, the link is checked; if a marker fails, the speed is reduced or the machine stops, and a lock axis request is output; and a differentiated lock axis step, where lock axis mappings are configured according to machine type, locking only the axes that change the horizontal outer envelope, and the lock axis action is output by merging the two request steps. This invention, through dual-channel redundant interlocking, dynamic safety distance prediction and determination, and differentiated lock axis, reliably prevents collisions and avoids secondary accidents caused by a unified complete shutdown even in the event of communication failure, thus balancing construction safety and continuity.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This application relates to the field of construction machinery control, and in particular to a method for the coordinated control of a group of civil construction machinery in a switchyard. Background Technology

[0002] Switch station civil construction sites are usually equipped with multiple types of mobile machinery such as rotary drilling rigs, mobile cranes, excavators, loaders and concrete mixer trucks. The working area is narrow and cross-operations are frequent. At the same time, the frequency converter of the pile driver generates broadband electromagnetic interference, the steel frame and the soil pile block the electromagnetic shadow area, and there is a multipath effect at the edge of the deep foundation pit. This results in a complex and unreliable wireless communication environment in the construction area, and status broadcast packet loss is common.

[0003] Safety in multi-machine collaborative operations is a critical issue in the civil construction of switchyards. Mechanical collisions can cause equipment damage, personnel injury, and construction interruption. However, maintaining construction continuity requires that interlocks cannot simply stop completely—the mixing drum of the tanker truck needs to rotate continuously to prevent segregation, the power head of the rotary drilling rig should not be locked for extended periods to prevent jamming, and the crane winch should not be suddenly locked to prevent the load from swinging. Therefore, it is necessary to establish an interlocking mechanism that balances collision prevention, construction continuity, and differences in mechanical kinematics.

[0004] Existing anti-collision interlocks have the following drawbacks: centralized interlocks rely on a central node, and when communication is unreliable, the judgment delay can reach the second level, and judgment based on expired coordinates will produce a positional deviation of several meters; schemes that rely solely on wireless communication will completely lose interlock input when communication continues to fail, and manual visual inspection has blind spots; uniformly "completely shut down" interlocking for all machinery will cause secondary accidents such as tanker rollaway, drill bit jamming, load swinging, and concrete segregation; and collision judgment based on single-frame coordinates cannot reflect the speed of mechanical approach, ignores the action envelope radius, and has unclear definitions of actuator response delay, resulting in inaccurate calculation of dynamic safety distance, which may misjudge collision risk as a safe state. Summary of the Invention

[0005] To address the problems of existing anti-collision interlocks relying on unreliable wireless communication, being prone to loss or misjudgment of interlocks when they fail, having blind spots in visual inspection, being prone to secondary accidents when all-around stops are implemented, and having difficulty accurately estimating safe distances with single-frame judgment, this invention provides solutions in the following aspects.

[0006] A collaborative control method for a group of civil construction machinery in a switchyard includes: a distributed interlocking determination step, whereby each machine acquires its planar position and action status, and broadcasts status information according to a status broadcast cycle, the status information including the machine's planar position and action status; receiving status information broadcast by neighboring machines and maintaining the status information of neighboring machines locally; determining the relative approach speed based on the planar position changes of the machine and neighboring machines in two adjacent frames, and constructing a dynamic safety distance by combining the total response delay of the actuators and the positioning error band; when the planar distance between the machine and neighboring machines is not greater than the sum of the action envelope radii of both machines and the dynamic safety distance, the local safety control unit of each machine outputs a shaft locking request; The magnetic induction backup verification step involves pre-burying passive magnetic induction markers in the construction area. When the machinery passes over the markers, the verification is triggered and the communication link status is determined. If the communication link fails, a conservative action of slowing down or stopping is performed and a shaft locking request is output. The differentiated locking axis step configures the locking axis mapping relationship according to the mechanical type, locks only the action axis that changes the horizontal outer envelope, and merges the locking axis request generated by the distributed interlock determination step and the magnetic induction backup verification step, and outputs the locking axis action according to the locking axis mapping relationship.

[0007] Preferably, the relative approach speed is: When the machine approaches a neighboring machine, the rate of decrease in the planar distance between two adjacent frames is taken; when they move away from each other, the relative approach speed is taken as zero. The planar distance between the two frames involved in the calculation is processed by a moving average of a preset duration. The time interval between two adjacent frames is determined by the difference between the timestamps of the status broadcast of the two frames. When the deviation of the time interval from the status broadcast period exceeds a preset ratio, the relative approach speed of the previous frame is used.

[0008] Preferably, the dynamic safety distance is: The speed estimation error band is obtained by taking twice the ratio of the positioning error band to the preset time. The dynamic safety distance is obtained by multiplying the sum of the relative approach speed and the speed estimation error band by the sum of the status broadcast period and the total delay of the actuator response, and adding the minimum static safety distance and the positioning error band.

[0009] Preferably, the triggering condition for the output lock axis request is: The planar distance between this machine and adjacent machinery is no greater than the sum of the machine's motion envelope radius, the adjacent machinery's motion envelope radius, and the dynamic safety distance; the motion envelope radius is determined by looking up a table according to the machinery type and its current motion status codeword. The motion envelope radius of the excavator is determined by the most unfavorable configuration corresponding to the maximum horizontal extension of its bucket boom.

[0010] Preferably, in the distributed interlock determination step, when no valid status broadcast of a nearby machine is received for multiple consecutive status broadcast cycles, the nearby machine is marked as having an unknown action status, and its action envelope radius is enlarged to the maximum action envelope radius of that machine type. At the same time, the dynamic safety distance is multiplied by a preset conservative coefficient, and the machine's travel speed is limited to not exceeding the low-speed passage limit. When the machine cannot obtain positioning information, the broadcast of valid coordinates is stopped until positioning is restored, during which time the machine's travel speed does not exceed the low-speed passage limit.

[0011] Preferably, in the magnetic induction backup verification step, the passive magnetic induction marker is composed of permanent magnets arranged in a predetermined polarity sequence, with different polarity sequences corresponding to different marker numbers; The mechanical chassis uses magnetic induction sensors on both sides to sense the marks it passes through, and determines the complete passage of the mechanical marks and the direction of passage based on the pulse sequence generated by the sensors on both sides in turn. The markers are divided into ordinary travel markers and cross risk markers according to the nature of the risk. The cross risk markers are arranged outside the cross operation boundary, and the arrangement distance is the larger value between the distance calculated by the upper limit of the trigger check speed and the mechanical braking capacity and the preset benchmark distance. The machine has a pre-stored marker layout table corresponding to the marker. The marker layout table records the number, type, expected location or expected interval of adjacent markers for each marker. When the continuous driving distance exceeds the expected interval and no marker is triggered, it is determined to be a marker-missing path. The machine is then treated as having a communication link failure and an unknown operating status in the cross-risk area ahead, and its speed is reduced to the low-speed passage limit.

[0012] Preferably, in the magnetic induction backup verification step, the verification includes speed constraints and data freshness determination: when the current driving speed of the machine is higher than the upper limit of the speed limit for triggering verification, it is first decelerated to the upper limit; when the interval between the most recent valid broadcast time of a neighboring machine and the current time exceeds the freshness threshold, the communication link is determined to be invalid and the area is marked as a communication interruption interval; within the communication interruption interval, the machine's driving speed does not exceed the low-speed passage limit, and each machine within the interval is treated as having an unknown operating status until a valid status broadcast is received again or the machine leaves the interval.

[0013] Preferably, when a communication link failure is determined, the machinery that has passed the cross-risk mark stops completely and observes for a predetermined period of time before passing through the marked area at a low speed limit, and continuously broadcasts its own status during the passage; the machinery that has passed the normal travel mark directly reduces its speed to the low speed limit, and the remaining machinery is treated as having an unknown operating status.

[0014] Preferably, the lock axis mapping relationship is as follows: Rotary drilling rigs lock the travel axis and slewing axis, while retaining the drilling or lifting axis; mobile cranes lock the travel axis, boom slewing axis, and winch lifting axis, while retaining outrigger extension and counterweight adjustment; excavators lock the travel axis and upper carriage slewing axis, while retaining bucket retraction and extension and stick pushing; loaders lock the travel axis and bucket lifting axis, while retaining the steering axis and bucket tilting axis; concrete mixer trucks lock the travel axis, and the mixing drum rotation is unrestricted; for any auxiliary axis movement that increases the horizontal outer envelope, its limit horizontal projection is included in the corresponding machine's motion envelope radius.

[0015] Preferably, the merging of the locking axis requests is a set union, that is, all action axes requested to be locked by different trigger channels are included in the final locking axis action; the locking axis release adopts the intersection determination, and the axis is released only when all trigger channels no longer request to lock a certain action axis; both the locking axis action and the release action are executed through a gradual ramp transition.

[0016] This application has the following technical advantages: 1. This invention achieves distributed interlock determination through local safety control units of each machine based on status broadcasts, and sets up a magnetic induction backup verification channel independent of the wireless communication link: the passive magnetic induction marker does not require power supply and does not rely on the communication link. The machine triggers the verification when it passes the marker. When the communication link fails, it can still perform conservative actions such as deceleration or stopping and output a shaft locking request. Therefore, this method can still maintain the interlock function under the condition that the wireless communication is unreliable or even that the local area is continuously failing. It overcomes the defects of the existing centralized interlock, which completely loses the interlock input when the communication fails and produces a position deviation of several meters based on the expired coordinates. It also makes up for the blind spots formed by the obstruction of soil piles and steel frames in manual visual inspection.

[0017] 2. This invention configures the locking axis mapping relationship according to the type of machinery, locking only the action axis that changes the horizontal outer envelope, while allowing the working axis that does not change the horizontal outer envelope to continue working. For example, rotary drilling rigs retain the drilling or lifting axis to prevent the drill bit from getting stuck in the hole; concrete mixer trucks retain the rotation of the mixing drum to prevent concrete segregation; mobile cranes retain the extension and retraction of outriggers and the adjustment of counterweight; loaders retain the steering and bucket tilting; and excavators retain the bucket extension and retraction and the stick pushing. Furthermore, based on the action envelope radius, it is ensured that the auxiliary axis actions that can continue do not exceed the envelope range that has been included in the judgment. Therefore, this method can maintain the continuity of construction while avoiding collisions between machinery, and avoid secondary accidents such as mixer truck slippage on ramps, drill bit jamming, load swinging, and concrete segregation caused by a complete shutdown.

[0018] 3. This invention determines the relative approach speed by the change in planar position between two adjacent frames. The planar distance between the two frames involved in the calculation is processed by a moving average of a preset duration, and the difference between the timestamps of the status broadcasts of the two frames is used as the time reference to suppress the speed estimation deviation caused by differential amplification of single-frame positioning errors and asynchronous broadcast time. Furthermore, the sum of the relative approach speed and the speed estimation error band is multiplied by the sum of the status broadcast period and the total delay of the actuator response, and the minimum static safety distance and the positioning error band are added to construct the dynamic safety distance. The value of the action envelope radius is obtained by looking up a table according to the machine type and action status codeword, and the value of the excavator is obtained according to the most unfavorable configuration. Therefore, this method can predict the machine approach risk, accurately estimate the dynamic safety distance, and avoid misjudging the collision risk as a safe state. Attached Figure Description

[0019] Figure 1 This is a flowchart of steps S1-S3 in the collaborative control method of the switch station civil construction machinery group in the embodiment of the present invention. Detailed Implementation

[0020] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are some embodiments of the present invention, but not all embodiments.

[0021] Reference Figure 1 The collaborative control method for the switchyard civil construction machinery group includes steps S1-S3, as detailed below: S1: Distributed interlock determination steps: Each machine obtains its own planar position and action status, and broadcasts status information according to the status broadcast cycle. The status information includes the machine's planar position and action status. It receives the status information broadcast by neighboring machines and maintains the status information of neighboring machines locally. It determines the relative approach speed based on the planar position changes of the machine and neighboring machines in two adjacent frames, and constructs a dynamic safety distance by combining the total response delay of the actuator and the positioning error band. When the planar distance between the machine and neighboring machines is not greater than the sum of the action envelope radius of both machines and the dynamic safety distance, the local safety control unit of each machine outputs a shaft locking request.

[0022] Each machine maintains a local dynamic catalog of neighboring machines. Each entry in the catalog records at least: the unique identifier of the neighboring machine, machine type code, action status codeword, planar coordinates, status broadcast timestamp, and count of consecutive missing broadcast cycles.

[0023] Each machine acquires its planar position and operational status, and broadcasts this status information according to a status broadcast cycle. The status broadcast cycle is the time interval between two consecutive status broadcasts, preferably 50 milliseconds in this embodiment. This cycle is set based on the principle that within a single status broadcast cycle, the horizontal displacement of the machine due to movement and rotation should be much smaller than the dynamic safety distance, thus ensuring that the coordinate data used for interlock determination is sufficiently up-to-date. If the status broadcast cycle is set too small, the wireless channel occupancy rate increases significantly, the probability of broadcast frame conflicts increases, and the machine's power consumption also rises. If the status broadcast cycle is set too large, the status updates of neighboring machines lag, the dynamic safety distance is calculated based on relatively outdated data, the perception delay increases, and the risk of missed collision detection increases. Considering all factors, the status broadcast cycle can be selected within the range of 20 to 200 milliseconds; this embodiment uses 50 milliseconds, a value that matches the output frequency of the satellite positioning module and the response delay of the actuator.

[0024] The action status codeword uses a four-bit status bitmap to sequentially express the occupancy status of the traveling axis, the occupancy status of the slewing axis, the occupancy status of the lifting or hoisting axis, and the rotation status of the stirring axis. This avoids semantic ambiguity caused by inconsistent descriptions of the same action status by different machines. Whenever a status broadcast cycle ends without receiving a broadcast frame from a neighboring machine, the consecutive missing broadcast cycle count for that neighboring machine in the catalog table is incremented by one. Once a valid broadcast frame is received, the count is immediately reset to zero, and the corresponding entry in the catalog table is updated with the newly received planar coordinates and the status broadcast timestamp.

[0025] Interlock determination is not a simple comparison of current coordinates, but a predictive determination based on relative proximity speed. The process for determining relative proximity speed is as follows: first, calculate the planar distance between the local machine and the neighboring machine at the current moment; then, calculate the planar distance between the local machine and the neighboring machine at the moment of the previous state broadcast cycle; subtract the current planar distance from the previous cycle's planar distance, and divide the difference by the time interval between two adjacent frames to obtain the relative proximity speed. The planar distances of the two frames involved in the calculation are first processed using a moving average.

[0026] The reason for introducing the moving average processing is that the positioning error of a single frame is amplified after being differentiated between two frames, which will seriously pollute the velocity estimation. If the coordinate difference of a single frame is used directly, the velocity error can reach several meters per second in the worst case, causing even stationary machines to appear to be approaching at high speed. The moving average window duration is the length of time covered by the moving average processing. In this implementation, it is set to 0.2 seconds, which is 4 state broadcast cycles. If the moving average window duration is set too small, the smoothing effect on positioning noise is insufficient, and the velocity estimation still carries obvious positioning jitter, causing the two machines to be misjudged as approaching each other when they are relatively stationary. If the moving average window duration is set too large, the response to the actual velocity change is lagging, and the velocity estimation value is too small when the actual approach speed of the machines is fast, resulting in insufficient dynamic safety distance. The moving average window duration can be selected within the range of 2 to 8 state broadcast cycles.

[0027] The time interval between two adjacent frames is determined by the difference between the status broadcast timestamps of the two frames. Since each machine broadcasts and receives data asynchronously, the actual time interval between two frames is not always equal to the nominal status broadcast period. Calculating it uniformly according to the nominal period would introduce additional errors. When the deviation between the status broadcast timestamps of two frames and the nominal status broadcast period exceeds a preset ratio, the time reference is considered abnormal, and the current frame does not update the relatively close speed, instead using the valid estimate from the previous frame. This preset ratio is set to distinguish between normal time jitter and abnormal time references: if the preset ratio is set too small, normal time reference fluctuations will also be judged as abnormal, causing speed estimates to frequently use old values ​​and lag behind; if the preset ratio is set too large, abnormal time references will be accepted as valid values, causing speed estimates to be incorrectly stretched or compressed. This preset ratio can be selected within the range of 20% to 50%, and this implementation uses 30%.

[0028] The relative approach speed adopts a signed rule: it is positive only when the machine and the adjacent machine are approaching each other, and the value is equal to the rate of decrease in the planar distance between two adjacent frames; when the machine and the adjacent machine are moving away from each other, the relative approach speed is zero, and the speed term no longer amplifies the dynamic safety distance, thus avoiding unnecessary shaft locking actions caused by false alarms in separation conditions. This rule overcomes the defect that when only the displacement of the adjacent machine is used, the speed calculation is zero when the machine actively approaches the stationary neighbor: in the three typical approach scenarios of the machine actively approaching the stationary neighbor, the neighbor actively approaching the machine, and both moving towards each other, the relative approach speed is greater than zero, and the interlock judgment can correctly reflect the approach risk.

[0029] After obtaining the relative approach speed, a dynamic safety distance is constructed: the sum of the relative approach speed and the speed estimation error band is multiplied by the sum of the status broadcast period and the total response delay of the actuator, and then the minimum static safety distance and the positioning error band are added. The result is the dynamic safety distance. The dynamic safety distance represents the minimum horizontal distance that should be maintained between the machine and the adjacent machine under the current approach speed and response delay conditions. It unifies speed, time delay, and positioning error into a distance quantity, covering all position changes from data acquisition to the actual braking of the actuator.

[0030] The velocity estimation error band reflects the velocity estimation uncertainty remaining after the moving average processing. It is calculated by dividing twice the positioning error band by the moving average window duration. This formula transmits the impact of positioning noise on the safe distance in the inverse order of the moving average window duration, rather than in the inverse order of the state broadcast period, thus avoiding false high-speed approach judgments caused by positioning jitter when the two machines are relatively stationary.

[0031] The positioning error band reflects the coordinate uncertainty under the multipath effect at the edge of the deep foundation pit. In this implementation, a historical statistical value of 0.2 meters is used. The positioning error band is set based on the measured statistical results of satellite positioning in the multipath environment of the deep foundation pit. If the positioning error band is set too small, the actual positioning deviation exceeds the assumed value, the dynamic safety distance is insufficient, and there is a risk of missed judgment. If the positioning error band is set too large, the dynamic safety distance is excessively amplified, the locking shaft is triggered too frequently, and normal operation is affected. The positioning error band can be selected within the range of 0.1 meters to 0.5 meters according to the accuracy of the positioning module and the site environment. In this implementation, 0.2 meters is used.

[0032] The total response delay of the actuator is defined as the total delay from the establishment of the interlock determination and the output of the locking shaft command to the hydraulic pilot cut-off, completion of the locking ramp, and complete cessation of the actuating shaft. It consists of the hydraulic pilot solenoid valve cut-off response time, the hydraulic circuit lag time, and the gradual locking ramp completion time. Its value is obtained through step response testing of the same model of mechanical actuator. In this embodiment, the typical value for the hydraulic pilot solenoid valve cut-off response time is 0.15 seconds, and the gradual locking ramp time is 0.5 seconds. The actual calculation is based on the total delay calibration value for this model. If the total response delay of the actuator is set too small, the actual braking has not yet been completed before the safety clearance is calculated based on a shorter delay, resulting in insufficient clearance. If the setting is too large, the safety clearance is amplified, leading to premature and frequent locking shaft triggering. The total response delay of the actuator should be calibrated separately according to the actual characteristics of each mechanical actuator and should not be used interchangeably across different models.

[0033] The minimum static safety distance is the minimum safety distance value corresponding to the type of machinery and is independent of speed. It is obtained by looking up a table in the safety database according to the model of the adjacent machinery. It is used to cover fixed factors such as the size of the machinery body, parking inertia and positioning residual deviation. Its specific value is related to the type of machinery and operating conditions and is configured according to the safety database.

[0034] When determining interlock based on the current planar distance, both the machine's motion envelope radius and the motion envelope radius of neighboring machines are considered. The interlock determination condition is: the planar distance between the machine and the neighboring machine is not greater than the sum of the machine's motion envelope radius, the neighboring machine's motion envelope radius, and the dynamic safety distance. The machine's motion envelope radius is determined by looking up a table using the machine type and current motion status codeword, while the neighboring machine's motion envelope radius is determined by looking up a table using the neighboring machine type and the most recently broadcast motion status codeword. The motion envelope radius represents the maximum horizontal range that the machine may occupy under the current motion configuration.

[0035] The motion envelope radius is determined by looking up a table based on the machine model and motion code. For example, when a rotary drilling rig is in rotation, its motion envelope radius includes not only the geometric radius of the rotating platform but also the range of drill rod sway. When a mobile crane boom rotates, the motion envelope radius is the maximum projection of the boom length onto the horizontal plane. When an excavator's upper structure rotates, the motion envelope radius is the outermost edge of the upper structure, including the projected radius of the bucket on the horizontal plane. When a loader is lifting, the motion envelope radius is the horizontal projection distance of the foremost edge of the bucket after lifting. In this embodiment, the motion envelope radius of the rotary drilling rig in rotation is 6.0 meters.

[0036] It is important to note that while the excavator's stick thrust significantly increases the horizontal projection at the front end, it does not remain unchanged in terms of the horizontal outer envelope. Therefore, this implementation establishes the following general principle for determining the motion envelope: For any auxiliary shaft movement that may increase the mechanical horizontal outer envelope, its limiting horizontal projection must be included in the lookup value of the motion envelope radius for that machine model; auxiliary shafts permitted to continue moving during interlocking must not cause the actual horizontal outer envelope to exceed the envelope value used in the interlock determination. For excavators, the lookup value for the motion envelope radius is taken based on the most unfavorable configuration with the bucket stick at its maximum horizontal extension, including the bucket, thus ensuring that the auxiliary shaft movements permitted to continue during interlocking do not exceed the envelope range already included in the determination.

[0037] When the interlock condition is met, the machine does not generate any remote control message. Instead, the local safety control unit directly de-energizes the corresponding hydraulic pilot solenoid valve, cuts off the control oil circuit of the corresponding actuating axis, and outputs a lock axis request. This hard-wired output does not depend on the communication link, so once the locking condition is met, the locked state can be maintained even if the subsequent broadcast is interrupted.

[0038] To address the issue of expired broadcast data, this step introduces a status freshness fallback rule: if a neighboring machine fails to update its catalog entry for multiple consecutive status broadcast cycles, the neighboring machine is marked as having an unknown action status, and its action envelope radius is temporarily enlarged to the maximum action envelope radius of that machine type, i.e., covering the maximum horizontal projection corresponding to all action status codes of that machine type. At the same time, the dynamic safety distance is multiplied by a conservative coefficient, and the machine's travel speed is limited to not exceeding the low-speed passage limit.

[0039] The threshold for consecutive missing broadcast cycles, i.e., the number of consecutive missing broadcast cycles required to determine an unknown operation status, is set based on the principle that a brief, single packet loss should not trigger an unknown status; only continuous failure should trigger conservative processing. If this threshold is set too low, even occasional packet loss will trigger an unknown status, causing the machine to frequently enter conservative operating mode, affecting the continuity of operations. If this threshold is set too high, genuine communication failures will not be identified for extended periods, and the machine will continue to make judgments based on outdated coordinates, increasing the risk of missed detections. This threshold can be selected within the range of 3 to 10 status broadcast cycles; this implementation uses 5 status broadcast cycles.

[0040] The conservatism coefficient is used to amplify the dynamic safety clearance when the state is unknown, in order to cover the additional risks brought about by the uncertainty of the motion state of adjacent machinery. If the conservatism coefficient is set too small, there is insufficient safety margin, and the amplified safety clearance is not much different from the normal state, which cannot effectively cover unknown motion; if the conservatism coefficient is set too large, the dynamic safety clearance is excessively amplified, the locking shaft is triggered frequently, and the operation efficiency decreases. The conservatism coefficient can be selected in the range of 1.2 to 2.0, and 1.5 is used in this implementation.

[0041] The low-speed limit is the upper speed limit under conservative driving conditions. It is set to ensure that the driver has sufficient time for visual confirmation and to implement emergency braking. If the low-speed limit is set too low, mechanical traffic efficiency will be too low, affecting construction progress; if the low-speed limit is set too high, braking distance will be insufficient, rendering conservative actions meaningless. The low-speed limit can be selected within the range of 1 km / h to 5 km / h; in this implementation, 2 km / h is used.

[0042] When the satellite positioning module is unable to locate for an extended period, such as when it is completely obscured by a deep foundation pit, the machine stops broadcasting valid coordinates. Nearby machinery treats the machine as having an unknown operational status. During the period when it cannot locate, the machine's speed is also limited to a low-speed passage limit until positioning is restored. This fallback logic ensures that when communication fails, it will not mistakenly allow passage due to missing data, but will instead shift in a conservative direction to avoid missed judgments caused by using outdated coordinates.

[0043] S2: Magnetic induction backup verification step. Passive magnetic induction markers are pre-embedded in the construction area. When the machinery passes over the markers, the verification is triggered and the communication link status is determined. When the communication link fails, conservative actions such as speed reduction or stopping are performed and a shaft locking request is output.

[0044] For objective scenarios where wireless communication fails continuously in local areas, such as when the frequency converter of a pile driver generates broadband electromagnetic interference during the peak drilling current period, and the temporary steel frame, high-density soil pile, and reflectors at the edge of the mud pit together form an electromagnetic shadow area, once the machinery enters such an area, short-range communication may be unable to complete effective frame exchange for several seconds, this step adopts the method of pre-embedding passive magnetic induction markers to provide a physical triggering channel for communication failure states that does not rely on power supply and wireless link.

[0045] The pre-embedded locations of the magnetic induction markers are determined synchronously with the civil construction process: passive magnetic induction markers are pre-embedded within the curbs of temporary roads, at the edges of pile cross casings, and at the corners of foundation pit ramps. Each marker carries a unique number. The physical implementation of the markers is as follows: several permanent magnets are arranged in a predetermined polarity sequence, with different polarity sequences corresponding to different marker numbers; a magnetic induction sensor is installed on each of the left and right sides of the mechanical chassis, using magnetoresistive sensors or Hall sensors to sense changes in the magnetic field and output corresponding pulse sequences; the local safety control unit decodes the marker number based on the polarity, quantity, and time interval of the pulses; when the left and right sensors generate valid pulses sequentially, it is determined that the machinery has completely passed a marker, and the sequence of the left and right differential pulses is used to distinguish the direction of passage, while suppressing false triggering caused by unilateral electromagnetic interference. The magnetic induction sampling adopts an event-triggered method, does not generate periodic power consumption, and the passive markers do not require power supply, making them suitable for unattended maintenance throughout the entire life cycle of civil construction sites.

[0046] The markers are categorized into ordinary travel markers and cross-risk markers based on their risk nature. Cross-risk markers are placed outside the cross-operation boundary, and the placement distance should meet the following requirements: not less than the square of the upper limit of the trigger verification speed divided by twice the mechanical braking deceleration, plus the upper limit of the trigger verification speed multiplied by the status broadcast period and the total delay of the actuator response. In actual placement, the larger of the distance calculated above and the preset reference distance is taken and registered in the marker layout table. This placement distance ensures that when the machinery travels to the cross-operation boundary at the upper limit of the trigger verification speed, there is sufficient distance to complete deceleration and verification. The preset reference distance in this implementation is 1.5 meters, used to provide a minimum placement limit for low-speed scenarios.

[0047] The upper limit for triggering the verification speed is set lower than the normal operating speed of the machinery, ensuring a smooth transition to the verification state at the marked location. If the upper limit is set too low, the machinery will frequently trigger verification during normal operation, affecting work efficiency; if set too high, the machinery will not have enough deceleration distance when reaching the marked location at high speed, making it impossible to complete the verification at the marked location. The upper limit for triggering the verification speed can be selected from 50% to 80% of the normal operating speed of the machinery. In this embodiment, it is configured in the local safety control unit according to the normal operating speed of each machine model.

[0048] Upon magnetic induction triggering, a status check is forcibly performed regardless of whether wireless communication is normal. The check process is divided into a speed constraint stage and a data freshness determination stage.

[0049] In the speed constraint process, if the current travel speed of the machine is higher than the upper limit of the speed limit for triggering the check, it will automatically decelerate to the upper limit value. The travel speed is taken from the chassis wheel speed or encoder odometer, or fused with satellite positioning speed, and does not rely solely on positioning calculation, thus ensuring that the speed constraint remains effective when positioning fails.

[0050] In the data freshness determination process, the most recent valid broadcast timestamp of each neighboring machine in the local catalog table is read. If the interval between the most recent valid broadcast time and the current time does not exceed the state freshness threshold, the communication link is considered healthy, and the communication data is directly used for verification. If it exceeds the state freshness threshold, the communication link is considered to have failed, and the local safety control unit marks the area as a communication interruption interval. The state freshness threshold is set to match the state broadcast cycle to distinguish between communication health and failure. If the state freshness threshold is set too low, even brief communication jitter will be judged as failure, and the machine will frequently enter a conservative state. If it is set too high, expired broadcast data will still be accepted, and the data freshness determination will lose its meaning. The state freshness threshold can be set to 3 to 10 state broadcast cycles; this implementation uses 5 state broadcast cycles.

[0051] The conservative action within the communication interruption interval is uniformly defined as follows: all adjacent machines within the interval are treated as having unknown action status, that is: the action envelope radius is enlarged to the maximum action envelope radius of the machine model, the dynamic safety distance is multiplied by the conservative coefficient, the machine's speed does not exceed the low-speed passage limit, until a valid broadcast from any adjacent machine is received again after two consecutive status broadcast cycles, or the machine leaves the interval.

[0052] For cross-risk markings, if the communication link has failed, a forced standby check is performed: the machinery completely stops and is observed for a predetermined time, during which time it continues to attempt to receive status broadcasts; if no valid status broadcast is received after the predetermined time, the machinery slowly passes through the marked area at a low speed limit, continuously broadcasting its own status during the passage. The observation time is set based on a balance between the duration characteristics of electromagnetic interference and construction progress requirements. If the observation time is set too short, the machinery may be released before the electromagnetic interference dissipates, potentially causing the check to fail; if it is set too long, the machinery may remain at the marked location for too long, affecting the construction progress. The observation time can be selected within the range of 1 to 5 seconds; this implementation uses 2 seconds.

[0053] For ordinary traveling markers, if the communication link has failed, a forced stop check will not be performed, but the speed will still be reduced to the low-speed passage limit, and other nearby machines will be handled conservatively as if their operating status is unknown.

[0054] To avoid missed detections due to magnetic induction markers being covered by mud or contaminated by metal shavings, the machine records the distance traveled without triggering any markers. Each machine has a locally stored marker layout table, recording the marker number, marker type, expected location, or expected interval between adjacent markers and the type of area they belong to. If the continuous travel distance exceeds the threshold of the expected interval between adjacent markers in the marker layout table without triggering a marker, it is determined to be a marker-missing path. When a marker is missing, the local safety control unit identifies the upcoming intersection risk area based on the marker layout table and treats this area as a situation of communication interruption and the presence of unknown machinery, automatically reducing the speed to the low-speed passage limit. This fallback rule prevents the system from silently failing after a marker is damaged, ensuring that the machine can still travel conservatively even when a marker is missing.

[0055] Meanwhile, the lock axis request generated by the magnetic induction trigger verification is independent and parallel to the lock axis request generated in step one, and they are not dependent on each other; the magnetic induction channel does not rely on any wireless communication resources, and its trigger determination, speed constraint, data freshness determination and lock axis request output are all completed by the local security control unit.

[0056] S3: Differentiated axis locking step. Configure axis locking mapping relationship according to machine type. Only lock the action axis that changes the horizontal outer envelope. After merging the axis locking request generated by the distributed interlock determination step and the magnetic induction backup verification step, output the axis locking action according to the axis locking mapping relationship.

[0057] To address the conflict between differences in mechanical kinematics and the requirements of construction continuity, uniformly implementing complete shutdown interlocks for all machinery would create three derivative risks: First, concrete mixer trucks would face difficulties restarting after a complete stop on a foundation pit ramp, and ramp parking could lead to slippage. Second, if the power head of a rotary drilling rig is locked while drilling in a pile hole, the drill rod remaining in the hole could cause diameter reduction or jamming. Third, if the winch is suddenly locked during crane hoisting, the suspended load could sway, exacerbating the impact on the hoisting mechanism. Therefore, interlock actions must be based on the kinematic characteristics of the machine model, limiting the axes most likely to cause collisions while allowing axes unrelated to collisions to continue operating.

[0058] Interlocking targets four types of moving axes: traveling axes, slewing axes, lifting or hoisting axes, and mixing axes. In addition, there are auxiliary operating axes not included in interlocking, such as the bucket extension / retraction axis and stick push axis of an excavator, and the steering axis of a loader. These auxiliary operating axes only change the local operating configuration or direction of travel, without altering the overall horizontal envelope of the machine, or causing horizontal interference with adjacent machines.

[0059] In this embodiment, under the civil construction scenario of the switch station, when five types of machinery are simultaneously configured, including rotary drilling rigs, mobile cranes, excavators, loaders, and concrete mixer trucks, the following differentiated shaft locking mapping relationship is established: (1) Rotary drilling rig: When the interlock is triggered, the traveling axis and the slewing axis of the rotary platform are locked, but the drilling or lifting axis is not locked. The reason is that the drilling and lifting actions occur within the vertical projection plane of the pile hole, which does not change the horizontal projection of the outer envelope of the machine and will not interfere with adjacent machines; retaining the drilling action can also prevent the drill bit from getting stuck in the hole.

[0060] (2) Mobile crane: Lock the traveling shaft, boom slewing shaft and hoisting shaft, but do not lock the outrigger extension and counterweight adjustment actions.

[0061] (3) Excavator: Lock the travel axis and the upper swing axis, allowing the bucket extension and retraction and stick pushing actions to continue during the interlock period, provided that the action envelope radius is taken from the table value according to the maximum horizontal extension of the stick, including the most unfavorable configuration of the bucket, so that the auxiliary axis actions allowed to continue do not exceed the envelope range already included in the judgment.

[0062] (4) Loader: Lock the travel axis and bucket lifting axis, but do not lock the steering axis and bucket tilting axis to ensure that it can still unload the material in the bucket and reduce the load when waiting; the limit horizontal projection of the tilting action has been included in the loader action envelope radius lookup table value.

[0063] (5) Concrete mixer truck: Only the drive shaft is locked, and the rotation of the mixing drum is not restricted in any way, preventing concrete segregation caused by long-term parking; the rotation of the mixing drum does not generate an external collision envelope.

[0064] The lock shaft mapping relationship is stored in the local safety control unit as a lock shaft mapping matrix and preset through the configuration interface. The five types of machinery and their mapping relationship in this embodiment are the preferred configurations for the civil construction scenario of the switch station. In actual applications, they can be adjusted through the configuration interface according to the construction machinery configuration, which does not constitute a limitation on the types of machinery applicable to this application.

[0065] When S1 or S2 triggers a locking request, the local controller retrieves the corresponding locking axis control set from the mapping matrix based on the local machine type, forming the final execution signal. Locking requests from different triggering sources converge into the local interlock arbiter. Locking adopts a union rule: if the magnetic induction verification channel requests to lock the travel axis, while the communication determination channel requests to lock the rotation axis, then both the travel axis and the rotation axis are ultimately locked simultaneously, and the union of the two requests is taken. This rule ensures that any single channel failure will not weaken the final locking range.

[0066] The axis lock release uses an intersection decision rule: an axis is only allowed to be released if it is not in the lock request sets of both channels simultaneously; if either channel still requests to lock the axis, it remains locked. The release and lock decisions share the same arbitrator and progressive ramp mechanism to prevent premature release due to misjudgment by a single channel, which would weaken safety.

[0067] The execution signal is output via a hydraulic pilot cut-off method, supplemented by a progressive locking ramp: the transition from the unlocked state to full locking is via a ramp of a preset duration, preventing the actuator from locking instantly and causing mechanical shock; similarly, when the interlock condition disappears, the release action also resumes via a ramp, preventing sudden mechanical movement. The duration of this progressive ramp is set to balance the locking shaft establishment speed and the impact suppression of the actuator. If the ramp duration is set too short, the actuating shaft will lock instantly, causing hydraulic shock and mechanical movement; if it is set too long, the locking shaft will establish slowly, and the mechanism may continue to move a distance before locking. The progressive locking ramp duration can be selected within the range of 0.2 seconds to 1.0 seconds; in this embodiment, 0.5 seconds is used.

[0068] Both locking and releasing events are recorded in the local operation log unit, including the locked axis number, trigger source type (i.e., communication judgment channel or magnetic induction verification channel), trigger target machine identification number, locking timestamp, and release timestamp, for post-event analysis and accountability. Each machine's local safety control unit also receives updates to the locking axis mapping matrix via a configuration interface to adjust the locking axis range when construction procedures change; this configuration update process is independent of real-time interlock judgment and does not interrupt current operation.

[0069] It should be noted that those skilled in the art can make various modifications and improvements without departing from the inventive concept, and these all fall within the scope of protection of this invention. Therefore, the scope of protection of this patent should be determined by the appended claims.

Claims

1. A method for coordinated control of a group of civil construction machinery in a switchyard, characterized in that the method... include: The distributed interlock determination process involves each machine acquiring its own planar position and operational status, and broadcasting status information according to a status broadcast cycle. The status information includes the machine's planar position and operational status. It also receives status information broadcast by neighboring machines and maintains this information locally. The relative approach speed is determined based on the planar position changes of the machine and neighboring machines in two adjacent frames, and a dynamic safety distance is constructed by combining the total response delay of the actuators with the positioning error band. When the planar distance between the machine and neighboring machines is not greater than the sum of their operational envelope radii and the dynamic safety distance, the local safety control unit of each machine outputs a shaft locking request. The magnetic induction backup verification step involves pre-burying passive magnetic induction markers in the construction area. When the machinery passes over the markers, the verification is triggered and the communication link status is determined. If the communication link fails, a conservative action of slowing down or stopping is performed and a shaft locking request is output. The differentiated locking axis step configures the locking axis mapping relationship according to the mechanical type, locks only the action axis that changes the horizontal outer envelope, and merges the locking axis request generated by the distributed interlock determination step and the magnetic induction backup verification step, and outputs the locking axis action according to the locking axis mapping relationship.

2. The method for coordinated control of a group of civil construction machinery in a switchyard according to claim 1, characterized in that, The relative approach speed is: When the machine approaches a neighboring machine, the rate of decrease in the planar distance between two adjacent frames is taken; when they move away from each other, the relative approach speed is taken as zero. The planar distance between the two frames involved in the calculation is processed by a moving average of a preset duration. The time interval between two adjacent frames is determined by the difference between the timestamps of the status broadcast of the two frames. When the deviation of the time interval from the status broadcast period exceeds a preset ratio, the relative approach speed of the previous frame is used.

3. The method for coordinated control of a group of civil construction machinery in a switchyard according to claim 2, characterized in that, The dynamic safety distance is: The speed estimation error band is defined as twice the ratio of the positioning error band to the preset time. The dynamic safety distance is obtained by multiplying the sum of the relative approach speed and the speed estimation error band by the sum of the state broadcast period and the total delay of the actuator response, and then adding the minimum static safety distance and the positioning error band.

4. The method for coordinated control of a group of civil construction machinery in a switchyard according to claim 1, characterized in that, The triggering condition for the output lock axis request is: The planar distance between this machine and adjacent machinery is no greater than the sum of the machine's motion envelope radius, the adjacent machinery's motion envelope radius, and the dynamic safety distance; the motion envelope radius is determined by looking up a table according to the machinery type and its current motion status codeword. The motion envelope radius of the excavator is determined by the most unfavorable configuration corresponding to the maximum horizontal extension of its bucket boom.

5. The method for coordinated control of a group of civil construction machinery in a switchyard according to claim 1, characterized in that, In the distributed interlock determination step, when no valid status broadcast of a nearby machine is received for several consecutive status broadcast cycles, the nearby machine is marked as having an unknown action status, and its action envelope radius is enlarged to the maximum action envelope radius of that machine type. At the same time, the dynamic safety distance is multiplied by a preset conservative coefficient, and the machine's travel speed is limited to not exceeding the low-speed passage limit. When the machine cannot obtain positioning information, the broadcast of valid coordinates is stopped until positioning is restored, during which time the machine's travel speed does not exceed the low-speed passage limit.

6. The method for coordinated control of a group of civil construction machinery in a switchyard according to claim 5, characterized in that, In the magnetic induction backup verification step, the passive magnetic induction marker is composed of permanent magnets arranged in a predetermined polarity sequence, with different polarity sequences corresponding to different marker numbers; The mechanical chassis uses magnetic induction sensors on both sides to sense the marks it passes through, and determines the complete passage of the mechanical marks and the direction of passage based on the pulse sequence generated by the sensors on both sides in turn. The markers are divided into ordinary travel markers and cross risk markers according to the nature of the risk. The cross risk markers are arranged outside the cross operation boundary, and the arrangement distance is the larger value between the distance calculated by the upper limit of the trigger check speed and the mechanical braking capacity and the preset benchmark distance. The machine has a pre-stored marker layout table corresponding to the marker. The marker layout table records the number, type, expected location or expected interval of adjacent markers for each marker. When the continuous driving distance exceeds the expected interval and no marker is triggered, it is determined to be a marker-missing path. The machine is then treated as having a communication link failure and an unknown operating status in the cross-risk area ahead, and its speed is reduced to the low-speed passage limit.

7. The method for coordinated control of a group of civil construction machinery in a switchyard according to claim 6, characterized in that, In the magnetic induction backup verification step, the verification includes speed constraints and data freshness determination: when the current travel speed of the machine is higher than the upper limit of the speed at which the verification is triggered, it is first decelerated to the upper limit; When the interval between the most recent valid broadcast time of a nearby machine and the current time exceeds the freshness threshold, the communication link is determined to be invalid and the area is marked as a communication interruption interval. Within the communication interruption interval, the machine's speed does not exceed the low-speed passage limit, and each machine within the interval is treated as having an unknown operating status until it receives a valid status broadcast again or leaves the interval.

8. The method for coordinated control of a group of civil construction machinery in a switchyard according to claim 7, characterized in that, When a communication link failure is detected, the machinery that has passed through the cross-risk marker will stop completely and observe for a predetermined period of time before passing through the marked area at a low speed limit, and will continuously broadcast its own status during the passage; the machinery that has passed through the normal travel marker will directly reduce its speed to the low speed limit, and the remaining machinery will be treated as having an unknown operating status.

9. The method for coordinated control of a group of civil construction machinery in a switchyard according to claim 1, characterized in that, The lock axis mapping relationship is as follows: Rotary drilling rigs lock the travel axis and slewing axis, while retaining the drilling or lifting axis; mobile cranes lock the travel axis, boom slewing axis, and winch lifting axis, while retaining outrigger extension and counterweight adjustment; excavators lock the travel axis and upper carriage slewing axis, while retaining bucket retraction and extension and stick pushing; loaders lock the travel axis and bucket lifting axis, while retaining the steering axis and bucket tilting axis; concrete mixer trucks lock the travel axis, and the mixing drum rotation is unrestricted; for any auxiliary axis movement that increases the horizontal outer envelope, its limit horizontal projection is included in the corresponding machine's motion envelope radius.

10. The method for coordinated control of a group of civil construction machinery in a switchyard according to claim 1, characterized in that, The merging of the locking axis requests is a set union, that is, all action axes requested to be locked by different trigger channels are included in the final locking axis action; the locking axis release adopts the intersection determination, and the axis is released only when all trigger channels no longer request to lock a certain action axis; both the locking axis action and the release action are executed through a gradual ramp transition.