Three-dimensional space collision avoidance system based on multiple tower cranes

Abstract

The invention provides a three-dimensional space collision avoidance system based on multiple tower cranes, and relates to the technical field of port hoisting operation, a three-dimensional voxel grid VoxelGrid space map is constructed, and the three-dimensional space collision avoidance system comprises a first horizontal coverage map, a second descending coverage map, a third fan-shaped area occupation map, a fourth track occupation area map and a fifth interference area map; the system can identify the space overlapping area of the tower crane operation path at the voxel level, breaks through the limitation of the traditional path planning method on the dimension and precision of space modeling, and enhances the fine granularity and accuracy of collision detection. A dynamic acquisition module and a dynamic interference module are introduced, dynamic disturbance caused by factors such as stormy waves and tides on a ship in a port operation environment can be sensed in real time, the deviation trend of the tail end of the lifting appliance is predicted, target correction and track compensation are carried out before track execution, and the situation that the lifting appliance is accidentally touched or falls off due to track drifting is effectively promoted to be reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of port hoisting operations, and particularly to a three-dimensional space collision avoidance system based on multiple tower cranes. Background Art

[0002] Tower cranes usually refer to "shore bridges" or "port tower cranes" on the shore. They are not on the ship but installed on the quay side for loading and unloading operations on ships docked at the shore (such as containers, bulk cargo, heavy pieces, etc.); with the continuous improvement of port automation and intelligence levels, the efficiency of container or bulk cargo handling operations based on tower cranes (hereinafter referred to as "tower cranes") has been increasingly emphasized. In actual port loading and unloading operations, there are often situations where multiple shore-based tower cranes cooperate to operate on multiple hatch areas of the same large ship. Since the tower crane body structure is fixed to the shore base and does not have the ability to automatically adjust the operation trajectory with the movement of the ship, however, affected by factors such as tidal changes, wind and wave interference, and the ship's own swaying, there are small but continuous dynamic disturbances during berthing, causing the position of the operation object to shift spatially, resulting in a deviation between the original hoisting trajectory of the tower crane and the actual position of the ship's cabin, thus there is a potential collision risk.

[0003] Especially in the scenario of multi-tower crane cooperative operation, the operation areas between different tower cranes may overlap in three-dimensional space. If the ship disturbance cannot be sensed in time and the hoisting trajectory cannot be corrected, it is easy to cause interference and collision between spreaders and between spreaders and the ship's hull structure, which not only affects the operation efficiency but may also lead to safety accidents. In the prior art, although some systems can achieve tower crane path planning or simple anti-collision alarms, they generally lack the ability to sense and intelligently respond to ship dynamic disturbances, cannot dynamically adjust the operation path based on disturbance information, and have a low-dimensional modeling of spatial conflicts under multi-tower crane cooperative operation, lacking high-precision, voxel-level three-dimensional collision analysis capabilities.

[0004] The above information disclosed in the above background art section is only used to enhance the understanding of the background of the present disclosure, so it may include information that does not constitute the prior art known to those of ordinary skill in the art. Summary of the Invention

[0005] The purpose of the present invention is to provide a three-dimensional space collision avoidance system based on multiple tower cranes to solve the problems raised in the above background art.

[0006] To achieve the above purpose, the present invention provides the following technical solutions:

[0007] A three-dimensional space collision avoidance system based on multiple tower cranes, comprising:

[0008] The trajectory acquisition module is used to establish a three-dimensional spatial coordinate conversion model for the port operation area, uniformly convert the operating coordinates of each tower crane into a floating spatial coordinate system with the center of the i-th ship hatch as the reference, collect the first operating trajectory of the j-th tower crane corresponding to the i-th ship in real time and model it, construct and obtain a spatial map composed of a three-dimensional voxel grid VoxelGrid, and obtain the first horizontal coverage map, the second descending coverage map, the third fan-shaped area occupancy map, the fourth track occupancy area map, and the fifth interference area map;

[0009] The trajectory recognition module is used to perform spatial conflict detection between tower cranes based on the first horizontal coverage map, the second descending coverage map, the third fan-shaped area occupancy map, the fourth track occupancy area map, and the fifth interference area map, and identify the set of overlapping voxels with potential spatial interference between the j-th tower crane and the adjacent j+1-th tower crane at the t-th operation time. when This indicates that there is a voxel-level spatial overlap area at the operation time t, indicating that there is a potential spatial interference risk between the j-th tower crane and the adjacent j+1-th tower crane at the operation time t. The first warning instruction is generated, and the interference risk index between the j-th tower crane and the adjacent j+1-th tower crane at the operation time t is constructed. Then, the classification is carried out and the corresponding avoidance strategy is generated according to the classification;

[0010] Dynamic collection module, collects dynamic disturbance information of ships in port scenes and generates dynamic disturbance data sets;

[0011] The dynamic interference module is used to predict the offset of the j-th tower crane spreader end in three-dimensional space at the t-th operation time, based on the dynamic disturbance data set, corresponding to the i-th ship's scheduled alignment point. After making target correction compensation for the original planned position, a first compensation instruction is obtained;

[0012] The expansion voxel correction module is used to obtain the second corrected operating trajectory of the j-th tower crane after obtaining the first compensation instruction, and collect the deviation value between the second operating trajectory and the first operating trajectory to obtain the expansion voxel size △Vp j , used to expand the second operating trajectory of the j-th tower crane after correction, and obtain the second operating trajectory of the j-th tower crane in the floating space coordinate system The trajectory recognition module repeatedly adopts the second operating trajectory of the j-th tower crane in the floating space coordinate system after expansion Perform overlapping voxel detection and correction.

[0013] Furthermore, the trajectory acquisition module includes a hull floating reference point acquisition unit;

[0014] The hull floating reference point acquisition unit is used to obtain the three-dimensional spatial position and attitude angle information of the center of the k-th hatch of the i-th ship, so as to construct a floating space coordinate system based on WebGL, Unity or UE.

[0015] Furthermore, the trajectory acquisition module further includes a three-dimensional voxel network construction unit and a spatial map extraction unit;

[0016] The three-dimensional voxel network construction unit is used for the scenario where multiple tower cranes are operating on port ships simultaneously. It collects the spatial trajectories of the lifting appliances of each tower crane in the unified floating coordinate system, performs time parameterization modeling, and maps the trajectories of different operation processes to various spatial maps. Spatial collision calculation is carried out based on three-dimensional voxel grids. The specific steps are as follows:

[0017] S11. Model the first operation trajectory of the lifting appliance of each tower crane in the floating space coordinate system to obtain the first operation trajectory of the lifting appliances of the j-th tower crane and the adjacent (j + 1)-th tower crane in the floating space coordinate system and The expression is:

[0018]

[0019] where t represents the operation time, x j (t), y j (t), z j (t) represent the coordinates of the spatial position of the lifting appliance of the j-th tower crane in the X, Y, and Z directions at the operation time t;

[0020] x j+1 (t), y j+1 (t), z j+1 (t) represent the coordinates of the spatial position of the lifting appliance of the adjacent (j + 1)-th tower crane in the X, Y, and Z directions at the operation time t;

[0021] S12. The spatial map extraction unit is used to construct a spatial map, which is specifically composed of a three-dimensional voxel grid VoxelGrid, and specifically includes:

[0022] S121. Extract the horizontal displacement section of the lifting appliance from the operation trajectory of the lifting appliance of the j-th tower crane in the floating space coordinate system, and map the first operation trajectory of the lifting appliances of the j-th tower crane and the adjacent (j + 1)-th tower crane in the floating space coordinate system and to the corresponding trajectory segments of the three-dimensional voxel grid to obtain the first horizontal coverage map and generate the first horizontal occupancy mark;

[0023] S122. Extract the path of the lifting appliance of the j-th tower crane descending from the air to the cabin from the trajectory. Specifically, it is zj (t) represents the vertical height of the spreader of the j-th tower crane at the operation time t, and it is recognized that the spreader is in the lowering operation state during the corresponding time period. The recognition rule is: and the spreader speed is not equal to 0; it means that at the operation time t, the spreader of the j-th tower crane is moving downward; where dz j (t) represents the position change of the spreader of the j-th tower crane in the vertical direction at the operation time t, that is: z j (t + Δt) - z j (t), and dt represents the time interval between two moments t and t + Δt;

[0024] After recognizing that the spreader is in the lowering operation state, the lowering trajectory segment is obtained The expression is:

[0025]

[0026] where is the start time of the lowering operation, is the end time of the lowering operation;

[0027] The lowering trajectory segment is sampled at equal intervals offline to obtain the trajectory line segment set:

[0028]

[0029] where t1, t2,... t n represent n sampling moments, and the time interval of the lowering trajectory segment is discrete time points;

[0030] And a continuous line segment set is generated:

[0031]

[0032] where and Every two consecutive sampling points form a lowering line segment, indicating the spatial movement of the spreader during this period; (t k ) and (t k+1 ) represent two consecutive time points, with a fixed time interval difference, (t k ) represents the k-th second, and (t k+1 ) represents the (k + 1)-th second;

[0033] Segment is a mathematical object that describes the linear motion between two points, representing the path of the spreader between two points;

[0034] For each lowering line segment L k, using the 3D Bresenham line projection algorithm, map the line segment to a 3D voxel grid, obtain the second descending coverage map, and generate the second descending occupancy mark;

[0035] S123. Extract the horizontal rotational motion component from the operation trajectory of the spreader of the j-th tower crane in the floating space coordinate system. Set each segment of the rotational motion as each sector area using the time frame, and map each sector area to a 3D voxel grid to form the third sector area occupancy map and generate the third rotational occupancy mark;

[0036] S124. Extract the common track or track intersection area of the operation space from the operation trajectories of the spreaders of the j-th tower crane and the adjacent (j + 1)-th tower crane in the floating space coordinate system, and expand the track width to form the fourth track occupancy area map;

[0037] S125. Extract the 3D overlapping volume from the operation trajectories of the spreaders of the j-th tower crane and the adjacent (j + 1)-th tower crane in the floating space coordinate system to form the fifth interference area map.

[0038] Furthermore, the trajectory recognition module includes a conflict detection unit and a first avoidance unit;

[0039] The conflict detection unit is used to count the first horizontal coverage map, the second descending coverage map, the third sector area occupancy map, the fourth track occupancy area map, and the fifth interference area map in S121 - 125, and perform the following operations to achieve the detection of spatial conflicts between tower cranes:

[0040] S131. Extract the first horizontal coverage map, the second descending coverage map, the third sector area occupancy map, the fourth track occupancy area map, and the fifth interference area map corresponding to S121 - 125, and identify the overlapping voxel set with potential spatial interference between the j-th tower crane and the adjacent (j + 1)-th tower crane at the t operation moment The expression is:

[0041]

[0042] In the formula, represents the voxel set of the r-th type of occupancy map of the j-th tower crane at the t operation moment, where r = 1 is the first horizontal coverage map; r = 2 is the second descending coverage map; r = 3 is the third sector area occupancy map; r = 4 is the fourth track occupancy area map; r = 5 is the fifth interference area map; represents the voxel set of the r-th type of occupancy map of the adjacent (j + 1)-th tower crane at the t operation moment;

[0043] When Denotes an empty set, indicating that there is a spatial overlap region at the voxel level at the t operation moment, indicating that there is a potential spatial interference risk between the j-th tower crane and the adjacent (j + 1)-th tower crane at the t operation moment, and generating a first warning instruction; when Indicates that there is no potential spatial interference risk between the j-th tower crane and the adjacent (j + 1)-th tower crane, and there is no spatial overlap in the current operation trajectory at the t operation moment;

[0044] S132. After receiving the first warning instruction, extract the number of voxel elements in the overlapping voxel set where there is potential spatial interference between the j-th tower crane and the adjacent (j + 1)-th tower crane at the t operation moment in the overlapping area, and construct the interference risk index between the j-th tower crane and the adjacent (j + 1)-th tower crane at the t operation moment through the following formula

[0045]

[0046] In the formula, |V safe | represents the set number of voxel elements in the set maximum spatial redundancy safety area, and the result of GI j,j+1 obtains a normalized value in the [0, 1] interval, which is convenient for grading.

[0047] Furthermore, the first avoidance unit is used to grade the interference risk index value between the j-th tower crane and the adjacent (j + 1)-th tower crane at the t operation moment, including:

[0048] When it indicates that there is no intersection of the spatial operation voxels of the two tower cranes at the t operation moment, generating a first safety level, indicating that the j-th tower crane and the adjacent (j + 1)-th tower crane continue to operate normally;

[0049] When it indicates that there is an overlap in the edge area between the two tower cranes at the t operation moment, generating a second low-risk level strategy, including: recording the state and continuing to operate, and reducing the boom amplitude of the j-th tower crane by 3% - 5%, reducing the hook speed at the trolley position by 3% - 5%, adjusting the lifting height in the opposite direction of the overlap by 3% - 5% and smoothly reducing the overall speed by 10% - 20%;

[0050] When it indicates that the interference area enters the actual operation path of the tower arm / hook, with the potential for physical interference, and is marked as a conflict area, generating a third medium-risk level strategy, including: generating a second warning instruction, and offsetting 6% - 10% in the opposite direction of the overlap, reducing the boom amplitude of the j-th tower crane by 10% - 20%, reducing the hook speed at the trolley position by 6% - 10%, adjusting the lifting height in the opposite direction of the overlap by 6% - 10% and smoothly reducing the overall speed by 21% - 30%;

[0051] When it indicates that there is a collision risk between the two tower cranes at the t operation moment, and it is marked as a conflict area, generating a fourth-level high-risk strategy, including: generating a third warning instruction, adjusting the tower crane closer to the edge of the conflict area to give priority to avoidance, adjusting the boom amplitude of this tower crane to contract by ≥ 25%, immediately withdrawing from the conflict area, adjusting the lifting height by 30% in the opposite direction of overlap to get out of the dangerous intersection point, and smoothly reducing the speed by 40% - 50% as a whole; while the tower crane closer to the central axis suspends the lifting operation synchronously; then judge the interference risk index again until after the interference is released, the jth tower crane resumes operation through the main control intervention.

[0052] Furthermore, the dynamic disturbance data group includes: the roll angle pitch angle θ i,t of the ith ship corresponding to the spreader of the jth tower crane at the t operation moment, the tidal change height Δh tide,t and the wave interference amplitude A wave,t .

[0053] Furthermore, 7. The dynamic interference module includes an identification disturbance influence factor unit and a second avoidance unit;

[0054] The identification disturbance influence factor unit is used to extract the roll angle pitch angle θ i,t of the ith ship corresponding to the spreader of the jth tower crane at the t operation moment, the tidal change height Δh tide,t and the wave interference amplitude A wave,t from the dynamic disturbance data group. Because different tower cranes have different structures, different boom lengths and different stiffnesses, the disturbance influences are also different. As a shore-based fixed device, the structure of the tower crane will not "move by itself" due to the shaking of the ship, tides or waves. However, since the object of the tower crane's lifting operation, that is, the ship is in dynamic motion, it is necessary to "perceive these disturbances" and make intelligent response operations. It is necessary to establish a corresponding disturbance response coefficient matrix R j for the jth tower crane, and the expression is:

[0055]

[0056] where r 11 、r 12 、r 13 、r 14 are respectively the response sensitive offset values of the roll angle change to the X-direction offset, the response sensitive offset value of the pitch angle to the X-direction offset, the response sensitive offset value of the tidal change height to the X-direction offset, and the response sensitive offset value of the wave interference amplitude to the X-direction offset;

[0057] where r21 , r 22 , r 23 , r 24 are the sensitive offset values of the roll angle change to the Y-direction offset, the pitch angle to the Y-direction offset, the tidal change height to the Y-direction offset, and the wave interference amplitude to the Y-direction offset, respectively;

[0058] Among them, r 31 , r 32 , r 33 , r 34 are the sensitive offset values of the roll angle change to the Z-direction offset, the pitch angle to the Z-direction offset, the tidal change height to the Z-direction offset, and the wave interference amplitude to the Z-direction offset, respectively;

[0059] And establish the corresponding disturbance response coefficient matrix R according to the j-th tower crane j Convert it into a vector form, and predict and obtain the offset of the end of the spreader of the j-th tower crane in the three-dimensional space at the t-th operation moment relative to the i-th ship's predetermined alignment point for the corresponding operation The expression is:

[0060]

[0061] Among them, δx j,t , δy j,t and δz j,t are the response disturbance offsets of the end of the spreader of the j-th tower crane in the X, Y, and Z directions in the three-dimensional space after being affected by the shaking of the i-th ship at the t-th operation moment, respectively.

[0062] Furthermore, the second avoidance unit is used to perform target correction compensation on the offset of the end of the spreader of the j-th tower crane in the three-dimensional space at the t-th operation moment relative to the i-th ship's predetermined alignment point for the corresponding operation Specifically:

[0063] Set the original planned alignment point as:

[0064] Among them, and respectively represent the X, Y, and Z axis positions of the j-th tower crane relative to the i-th ship's predetermined alignment point;

[0065] Perform compensation on the offset of the end of the spreader of the j-th tower crane in the three-dimensional space at the t-th operation moment relative to the i-th ship's predetermined alignment point Perform compensation on the original planned alignment point to obtain the target correction position

[0066]

[0067] Among them, the target correction position This is the actual target position that the tower crane should move to after considering the ship disturbance; it is the new position obtained by adding the ship disturbance compensation to the original planned alignment point;

[0068] And a controller is used to convert the target correction position into the first compensation instruction, including: boom length adjustment amount, trolley displacement adjustment, hoisting height adjustment and speed adjustment.

[0069] Further, the inflated voxel correction module is used to obtain the second operation trajectory of the j-th tower crane after correction after obtaining the first compensation instruction, and collect and obtain the deviation value between the second operation trajectory and the first operation trajectory to obtain the inflated voxel size △Vp j for inflating the second operation trajectory of the j-th tower crane after correction to obtain the second operation trajectory of the j-th tower crane in the floating space coordinate system The inflated second operation trajectory will become a larger area, and the expression is:

[0070]

[0071] After obtaining the second operation trajectory of the j-th tower crane in the floating space coordinate system use the inflated second operation trajectory for overlapping voxel detection, and repeat S131 to S132 to reconstruct the interference risk index of the j-th tower crane and the adjacent (j + 1)-th tower crane at the t operation moment And classify and obtain the corresponding strategies and execute them.

[0072] Compared with the prior art, the beneficial effects of the present invention are as follows: By introducing a floating space coordinate system with the center of the ship's hatch as a reference, unified modeling of the operation trajectories of different onshore tower cranes under the same coordinate reference is realized, effectively eliminating the problem of spatial reference drift caused by ship swaying, and improving the accuracy and consistency of three-dimensional space perception during multi-tower crane collaborative operation. By constructing a three-dimensional voxel grid (VoxelGrid) spatial map and multiple types of space occupancy maps (such as the first horizontal coverage map, the second descending coverage map, the third fan-shaped area occupancy map, the fourth track occupancy area map, and the fifth interference area map), the system can identify the spatial overlapping areas of the tower crane operation paths at the voxel level, breaking through the limitations of traditional path planning methods in terms of spatial modeling dimension and accuracy, and enhancing the fine-grainedness and accuracy of collision detection.

[0073] The present invention introduces a dynamic acquisition module and a dynamic interference module, which can real-time sense the dynamic disturbances of ships caused by factors such as wind, waves, and tides in the port operation environment, predict the offset trend of the end of the spreader, and perform target correction and trajectory compensation before the trajectory execution, effectively promoting the reduction of spreader mis-collision or landing deviation caused by trajectory drift.

[0074] The present invention also uses the first avoidance unit to avoid the interference risk between tower cranes, and the second avoidance unit, that is: for the end of the spreader of the j-th tower crane in three-dimensional space at the t operation moment, the compensation for the offset vector corresponding to the position of the i-th ship during operation, that is: the dynamic compensation mechanism for the alignment error between the tower crane and the ship; the second avoidance unit calculates the offset of the end of the spreader of the j-th tower crane to obtain a new target correction position. Through this dynamic correction, the end of the tower crane spreader can be timely adjusted to an ideal position that conforms to the predetermined operation target, thus avoiding the alignment error caused by the dynamic interference of the ship. BRIEF DESCRIPTION OF THE DRAWINGS

[0075] Figure 1 It is a schematic flowchart of a three-dimensional space collision avoidance system based on multiple tower cranes of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0076] To make the objectives, technical solutions, and advantages of the present invention clearer and more understandable, the present invention will be further described in detail below with reference to specific embodiments.

[0077] It should be noted that, unless otherwise defined, the technical terms or scientific terms used in the present invention should have the ordinary meaning understood by those with ordinary skills in the field to which the present invention belongs. The "first", "second" and similar terms used in the present invention do not represent any order, quantity or importance, but are only used to distinguish different components. The terms such as "including" or "comprising" mean that the elements or objects appearing before this word cover the elements or objects listed after this word and their equivalents, without excluding other elements or objects. The terms such as "connected" or "coupled" are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. The terms such as "upper", "lower", "left", "right" are only used to represent relative position relationships, and when the absolute position of the object being described changes, the relative position relationship may also change accordingly.

[0078] Embodiment 1:

[0079] Please refer to Figure 1 , the present invention provides a technical solution:

[0080] A three-dimensional space collision avoidance system based on multiple tower cranes, comprising:

[0081] The trajectory acquisition module is used to establish a three-dimensional spatial coordinate conversion model for the port operation area, uniformly convert the operating coordinates of each tower crane into a floating spatial coordinate system with the center of the i-th ship hatch as the reference, collect the first operating trajectory of the j-th tower crane corresponding to the i-th ship in real time and model it, construct and obtain a spatial map composed of a three-dimensional voxel grid VoxelGrid, and obtain the first horizontal coverage map, the second descending coverage map, the third fan-shaped area occupancy map, the fourth track occupancy area map, and the fifth interference area map;

[0082] The trajectory recognition module is used to perform spatial conflict detection between tower cranes based on the first horizontal coverage map, the second descending coverage map, the third fan-shaped area occupancy map, the fourth track occupancy area map, and the fifth interference area map, and identify the set of overlapping voxels with potential spatial interference between the j-th tower crane and the adjacent j+1-th tower crane at the t-th operation time. when This indicates that there is a voxel-level spatial overlap area at the operation time t, indicating that there is a potential spatial interference risk between the j-th tower crane and the adjacent j+1-th tower crane at the operation time t. The first warning instruction is generated, and the interference risk index between the j-th tower crane and the adjacent j+1-th tower crane at the operation time t is constructed. Then, the classification is carried out and the corresponding avoidance strategy is generated according to the classification;

[0083] Dynamic collection module, collects dynamic disturbance information of ships in port scenes and generates dynamic disturbance data sets;

[0084] The dynamic interference module is used to predict the offset of the j-th tower crane spreader end in three-dimensional space at the t-th operation time, based on the dynamic disturbance data set, corresponding to the i-th ship's scheduled alignment point. After making target correction compensation for the original planned position, a first compensation instruction is obtained;

[0085] The expansion voxel correction module is used to obtain the second corrected operating trajectory of the j-th tower crane after obtaining the first compensation instruction, and collect the deviation value between the second operating trajectory and the first operating trajectory to obtain the expansion voxel size △Vp j , used to expand the second operating trajectory of the j-th tower crane after correction, and obtain the second operating trajectory of the j-th tower crane in the floating space coordinate system The trajectory recognition module repeatedly adopts the second operating trajectory of the j-th tower crane in the floating space coordinate system after expansion Perform overlapping voxel detection and correction.

[0086] In this embodiment, by introducing a floating space coordinate system with the center of the ship's hatch as the reference, the unified modeling of the operation trajectories of different onshore tower cranes under the same coordinate reference is realized, effectively eliminating the problem of spatial reference drift caused by ship swaying, and improving the accuracy and consistency of three-dimensional space perception during the collaborative operation of multiple tower cranes. By constructing a three-dimensional voxel grid (VoxelGrid) spatial map and multiple types of spatial occupancy maps (such as the first horizontal coverage map, the second descending coverage map, the third fan-shaped area occupancy map, the fourth track occupancy area map, and the fifth interference area map), the system can identify the spatial overlapping areas of the tower crane operation paths at the voxel level, breaking through the limitations of traditional path planning methods in terms of spatial modeling dimension and accuracy, and enhancing the fine-grainedness and accuracy of collision detection.

[0087] The system introduces a dynamic acquisition module and a dynamic interference module, which can real-time sense the dynamic disturbances caused by factors such as wind, waves, and tides on the ship in the port operation environment, predict the offset trend of the end of the spreader, and perform target correction and trajectory compensation before the trajectory execution, effectively avoiding the spreader mis-collision or landing deviation caused by trajectory drift.

[0088] By performing dilation processing on the corrected second operation trajectory, that is, the hoisting trajectory, and introducing the concept of "safety redundancy area", potential risk areas can be sensed in advance during the collision detection process and early warnings can be issued, further improving the anti-collision ability and robustness of the system. The system can not only identify the overlapping areas, but also construct the interference risk index between tower cranes at a specific operation moment and conduct risk level division based on this, so as to generate more targeted avoidance strategies, and have good intelligent scheduling and collaborative optimization capabilities.

[0089] Embodiment 2

[0090] This embodiment is an explanatory description based on Embodiment 1. Please refer to Figure 1 , specifically, the trajectory acquisition module includes a hull floating reference point acquisition unit;

[0091] The hull floating reference point acquisition unit is used to obtain the three-dimensional spatial position and attitude angle information of the center of the k-th hatch of the i-th ship, so as to construct a floating space coordinate system based on WebGL, Unity or UE.

[0092] Suppose there are 3 tower cranes that need to work together to unload the goods of a ship. The hull of the ship will change its height and attitude due to tides + loading and unloading swaying.

[0093] A coordinate system with the center of the k-th hatch of the i-th ship as the origin is established in the system, and the positions of all tower cranes, spreader trajectories, etc. need to be converted in real time. This "floating three-dimensional coordinate system" is actually a dynamic reference system model constructed and used, and it can also be integrated into a Web visualization system, a VR model, or an edge computing terminal.

[0094] In this embodiment, by collecting the spatial position and attitude angle information of the local hatch of the hull, during the operation path planning of the port shore-based tower crane, the drift, deflection, and sway of the ship itself can be synchronized in real time, making the constructed coordinate system more conform to the actual motion state of the hull, and effectively avoiding the problem of the disconnection between the path planning and the target operation point.

[0095] Embodiment 3

[0096] This embodiment is an explanatory description based on Embodiment 1. Please refer to Figure 1 , specifically, the trajectory acquisition module further includes a three-dimensional voxel network construction unit and a spatial atlas extraction unit;

[0097] The three-dimensional voxel network construction unit is used for the scenario where multiple tower cranes are operating on a port ship simultaneously. It collects the spatial trajectories of the spreaders of each tower crane in a unified floating coordinate system, performs time parameterization modeling, and maps the trajectories of different operation processes to various spatial atlases, and conducts spatial collision calculations based on three-dimensional voxel grids. The specific steps are as follows:

[0098] S11. Model the first operation trajectory of the spreader of each tower crane in the floating space coordinate system to obtain the first operation trajectories of the spreaders of the j-th tower crane and the adjacent (j + 1)-th tower crane in the floating space coordinate system and The expression is:

[0099]

[0100] where t represents the operation time, and x j (t), y j (t), z j (t) represent the coordinates of the spatial position of the spreader of the j-th tower crane in the X, Y, and Z directions at the operation time t; a set of horizontal voxel trajectories formed by horizontal projection line segments;

[0101] x j+1 (t), y j+1 (t), z j+1 (t) represent the coordinates of the spatial position of the spreader of the adjacent (j + 1)-th tower crane in the X, Y, and Z directions at the operation time t;

[0102] S12. The spatial map extraction unit is used to construct a spatial map, which is specifically composed of a three-dimensional voxel grid VoxelGrid, and specifically includes:

[0103] S121. Extract the horizontal displacement section of the spreader from the operation trajectory of the spreader of the j-th tower crane in the floating space coordinate system, and map the first operation trajectory of the spreaders of the j-th tower crane and the adjacent (j + 1)-th tower crane in the floating space coordinate system and the corresponding trajectory segments to the three-dimensional voxel grid to obtain the first horizontal coverage map and generate the first horizontal occupancy mark;

[0104] S122. Extract the path of the spreader of the j-th tower crane descending from the air to the cabin from the trajectory. Specifically, z j (t) represents the vertical height of the spreader of the j-th tower crane at the operation time t. During the corresponding time period, identify that the spreader is in the lowering operation state, and the identification rule is: and the spreader speed is not equal to 0; it means that at the operation time t, the spreader of the j-th tower crane is moving downward; where dz j (t) represents the position change of the spreader of the j-th tower crane in the vertical direction at the operation time t, that is: z j (t + Δt) - z j (t), and dt represents the time interval between two times t and t + Δt;

[0105] After identifying that the spreader is in the lowering operation state, obtain the lowering trajectory section The expression is:

[0106]

[0107] where, is the start time of the lowering operation, is the end time of the lowering operation;

[0108] Perform offline equidistant sampling on the lowering trajectory section to obtain a set of trajectory line segments:

[0109]

[0110] where, t1, t2,...t n represent n sampling times, and the time interval of the lowering trajectory section is discrete time points;

[0111] And generate a set of continuous line segments:

[0112]

[0113] where, and Every two consecutive sampling points form a lowering line segment, representing the spatial movement of the spreader during this period; (t k ) and (t k+1 ) represent two consecutive time points, with a fixed time interval between them. (t k ) represents the k-th second, and (t k+1 ) represents the (k + 1)-th second;

[0114] A Segment is a mathematical object describing the linear motion between two points, representing the path of the spreader between two points;

[0115] For each lowering line segment L k , using the 3D Bresenham line projection algorithm, map the line segment to a three-dimensional voxel grid to obtain the second lowering coverage map and generate the second lowering occupancy marker. A set of columnar voxel trajectories formed by projecting the line segment in the vertical direction;

[0116] S123. Extract the horizontal rotational motion component from the operation trajectory of the spreader of the j-th tower crane in the floating space coordinate system. Set each segment of rotational motion as each sector area with the time frame, and map each sector area to a three-dimensional voxel grid to form the third sector area occupancy map and generate the third rotational occupancy marker;

[0117] S124. Extract the common track or track intersection area of the operation space from the operation trajectories of the spreaders of the j-th tower crane and the adjacent (j + 1)-th tower crane in the floating space coordinate system, and expand the track width to form the fourth track occupancy area map;

[0118] S125. Extract the three-dimensional overlapping volume from the operation trajectories of the spreaders of the j-th tower crane and the adjacent (j + 1)-th tower crane in the floating space coordinate system to form the fifth interference area map.

[0119] In this embodiment, a three-dimensional voxel grid (VoxelGrid) is used as the basic unit for spatial analysis. By combining the dynamic trajectory modeling of the tower crane spreader in the floating space coordinate system, the spatial conflict detection and collision avoidance in the multi-tower crane operation scenario are realized. The time parameterized modeling can accurately capture the spatial position changes of the tower crane spreader at each operation moment, enhancing the real-time performance and accuracy of the spatial conflict analysis. The three-dimensional voxel network construction unit extracts various trajectory segments (such as horizontal displacement segments, descending path segments, and rotational motion components) through the precise modeling of the tower crane spreader operation trajectory, and maps them into the three-dimensional voxel grid, providing detailed spatial information for subsequent collision detection and path optimization. Especially in the multi-tower crane collaborative operation scenario, accurate analysis of the spatial overlapping area can be realized, effectively avoiding the interference between tower crane spreaders or between the spreader and the ship structure. Through the spatial map extraction unit, the system can generate multiple types of spatial maps, including the first horizontal coverage map, the second descending coverage map, the third sector area occupancy map, the fourth track occupancy area map, and the fifth interference area map. These maps can finely depict the working area of the tower crane according to different operation processes (such as spreader horizontal displacement, descending operation, rotation, etc.), providing multi-dimensional information support for spatial conflict detection during multi-tower crane operation, and significantly improving the overall operation efficiency and safety.

[0120] For the spreader lowering operation, the system monitors the vertical displacement and speed changes of the spreader, identifies and extracts the lowering trajectory of the spreader, and maps it into the three-dimensional voxel grid through the 3D Bresenham algorithm. This step ensures that the spatial conflicts during the spreader lowering process can be monitored in real time, avoiding the collision risk caused by ship disturbance or tower crane adjustment operations.

[0121] By extracting the rotational motion components in the horizontal direction of the tower crane operation trajectory, the system can form a sector area occupancy map and generate a track occupancy map by expanding the common track area. These strategies can effectively identify and avoid potential interference areas between tower cranes, especially when multiple tower cranes are operating on the same ship simultaneously, ensuring the effective allocation of the operation area and the rational use of space.

[0122] By real-time monitoring the operation trajectory of the tower crane spreader in the floating space coordinate system and timely generating an accurate interference area map, the system can accurately predict and detect potential collision risks. This makes the tower crane operation more intelligent and automated, effectively reducing the probability of accidents caused by human factors or ship disturbances, and improving the safety of port operations.

[0123] Embodiment 4

[0124] This embodiment is an explanatory note based on Embodiment 3. Please refer to Figure 1, specifically, the trajectory recognition module includes a conflict detection unit and a first avoidance unit;

[0125] The conflict detection unit is used to count the first horizontal coverage map, the second descending coverage map, the third fan-shaped area occupancy map, the fourth track occupancy area map, and the fifth interference area map in S121-125, and perform the following operations to achieve spatial conflict detection between tower cranes:

[0126] S131. Extract the first horizontal coverage map, the second descending coverage map, the third fan-shaped area occupancy map, the fourth track occupancy area map, and the fifth interference area map corresponding to S121-S125, and identify the overlapping voxel set with potential spatial interference between the j-th tower crane and the adjacent (j + 1)-th tower crane at the t operation moment The expression is:

[0127]

[0128] In the formula, represents the voxel set of the r-th occupancy map of the j-th tower crane at the t operation moment, where r = 1 is the first horizontal coverage map; r = 2 is the second descending coverage map; r = 3 is the third fan-shaped area occupancy map; r = 4 is the fourth track occupancy area map; r = 5 is the fifth interference area map; represents the voxel set of the r-th occupancy map of the adjacent (j + 1)-th tower crane at the t operation moment;

[0129] When represents an empty set, indicating that there is a spatial coincidence area at the voxel level at the t operation moment, indicating that there is a potential spatial interference risk between the j-th tower crane and the adjacent (j + 1)-th tower crane at the t operation moment, and a first warning instruction is generated; when represents that there is no potential spatial interference risk between the j-th tower crane and the adjacent (j + 1)-th tower crane, and there is no spatial coincidence in the current operation trajectory at the t operation moment;

[0130] S132. After receiving the first warning instruction, extract the number of voxel in the overlapping area of the overlapping voxel set with potential spatial interference between the j-th tower crane and the adjacent (j + 1)-th tower crane at the t operation moment in, and construct the interference risk index between the j-th tower crane and the adjacent (j + 1)-th tower crane at the t operation moment through the following formula

[0131]

[0132] In the formula, |V safe | represents the set number of voxels in the set maximum spatial redundancy safety area, and the result of GI j,j+1 gets a normalized value in the range of [0, 1], which is convenient for grading.

[0133] Example 5

[0134] This embodiment is explained in Example 4. Please refer to Figure 1 Specifically, the first avoidance unit is used to calculate the interference risk index between the j-th tower crane and the adjacent j+1-th tower crane at the operation time t. The values are graded, including:

[0135] when When , it means that at the operation time t, the spatial operation voxels of the two tower cranes have no intersection, generating the first safety level, indicating that the j-th tower crane and the adjacent j+1-th tower crane continue to operate normally;

[0136] when When , it means that there is an edge overlap between the two tower cranes at the operation time t, and the second lowest risk level strategy is generated, including: recording the status and continuing the operation, reducing the arm extension of the j-th tower crane by 3%-5%, lowering the trolley position hook speed by 3%-5%, adjusting the lifting height by 3%-5% in the opposite direction of the overlap, and smoothly reducing the overall speed by 10%-20%;

[0137] when When the interference area enters the actual operation path of the tower boom / spreader, there is a potential for physical interference and it is marked as a conflict area. The third medium risk level strategy is generated, including: generating a second early warning instruction, and offsetting the position in the opposite direction of overlap by 6%-10%, reducing the arm extension amplitude of the j-th tower crane by 10%-20%, reducing the trolley position hook speed by 6%-10%, adjusting the lifting height in the opposite direction of overlap by 6%-10%, and smoothly reducing the overall speed by 21%-30%;

[0138] when When , it means that there is a risk of collision between the two tower cranes at the operation time t, and it is marked as a conflict area, generating the fourth highest risk level strategy, including: generating the third warning instruction, adjusting the tower crane near the edge of the conflict area to give priority to avoidance, adjusting the arm of the tower crane to retract ≥ 25% of the arm extension, immediately withdrawing from the conflict area, adjusting the lifting height in the opposite direction of the overlap by 30%, leaving the dangerous intersection, and smoothly reducing the speed by 40%-50%; and the tower crane near the central axis will synchronously suspend the lifting operation; and then judge the interference risk index again. Until After the interference is lifted, the jth tower crane resumes operation through the intervention of the main control.

[0139] In this embodiment, by constructing a multi-dimensional atlas model of the tower crane operation trajectory, including the first horizontal coverage atlas, the second descending coverage atlas, the third fan-shaped area occupancy atlas, the fourth track occupancy area atlas, and the fifth interference area atlas, it can comprehensively reflect the occupancy behavior of the tower crane in different operation stages and spatial regions. Through the conflict detection unit in the trajectory recognition module, by performing intersection calculations on the voxel sets of the atlases of multiple tower cranes at the same operation moment, potential spatial conflict points can be quickly identified, avoiding the limitations of traditional methods that rely on single models or two-dimensional projections for judgment, and improving the accuracy and real-time performance of spatial interference recognition.

[0140] Furthermore, the "interference risk index" proposed in the embodiment is obtained through the normalization calculation between the number of overlapping voxels and the number of voxels in the maximum spatial redundancy safety area, making the interference risk degree between tower cranes quantifiable and comparable, facilitating system hierarchical processing, and supporting the execution of fine-grained safety control strategies.

[0141] On this basis, the first avoidance unit automatically generates differentiated avoidance strategies in combination with the classification of the risk index values, specifically including parameter regulation operations such as boom telescopic adjustment, trolley running speed reduction, hook lifting height correction, and overall running speed smooth reduction, enabling the tower crane to complete dynamic obstacle avoidance and spatial coordination without interrupting the operation, thereby significantly reducing the physical interference risk and collision probability in the multi-tower cooperation scenario.

[0142] In addition, this embodiment supports active warning and operation intervention for severe conflict situations (such as the interference area has entered the operation path of the tower arm / hook), realizes the priority avoidance of the tower crane close to the conflict edge and the synchronous pause of the remote tower crane through the main control system, and achieves the intelligent coordination and global scheduling of the tower crane operation process, ensuring the achievement of the dual goals of operation continuity and on-site safety.

[0143] Embodiment 6

[0144] This embodiment is an explanatory description based on Embodiment 54. Please refer to Figure 1 , specifically, the dynamic disturbance data group includes: the roll angle of the i-th ship corresponding to the hook of the j-th tower crane at the t operation moment pitch angle θ i,t , tidal change height Δh tide,t and wave interference amplitude A wave,t .

[0145] 7. The dynamic interference module includes an identification of disturbance influence factor unit and a second avoidance unit;

[0146] The identification of disturbance influence factor unit is used to extract the roll angle of the i-th ship corresponding to the hook of the j-th tower crane at the t operation moment from the dynamic disturbance data group Pitch angle Tidal change height Δh tide,t and wave interference amplitude A wave,t , because different tower cranes have different structures, different boom lengths, different stiffnesses, and different disturbance effects. As a shore-based fixed device, the structure of the tower crane will not "move by itself" due to the shaking of the ship, tides, or waves. However, since the object of the tower crane's hoisting operation, that is, the ship, is in dynamic motion, it is necessary to "perceive these disturbances" and thus make intelligent response operations. It is necessary to establish a corresponding disturbance response coefficient matrix R for the j-th tower crane j , and the expression is

[0147]

[0148] The following is the roll angle of the i-th ship corresponding to the operation Pitch angle Tidal change height Δh tide,t , wave interference amplitude A wave,t Example chart of the original factors affected

[0149]

[0150] Among them, r 11 , r 12 , r 13 , r 14 are the response sensitive offset values of the roll angle change to the X-direction offset, the pitch angle to the X-direction offset, the tidal change height to the X-direction offset, and the wave interference amplitude to the X-direction offset respectively;

[0151] Among them, r 21 , r 22 , r 23 , r 24 are the response sensitive offset values of the roll angle change to the Y-direction offset, the pitch angle to the Y-direction offset, the tidal change height to the Y-direction offset, and the wave interference amplitude to the Y-direction offset respectively;

[0152] Among them, r 31 , r 32 , r 33 , r 34 are the response sensitive offset values of the roll angle change to the Z-direction offset, the pitch angle to the Z-direction offset, the tidal change height to the Z-direction offset, and the wave interference amplitude to the Z-direction offset respectively;

[0153] And establish a corresponding disturbance response coefficient matrix R according to the j-th tower cranej Converted into vector form, predict and obtain the offset of the end of the spreader of the j-th tower crane in three-dimensional space corresponding to the i-th ship's predetermined alignment point at the t operation time. The expression is:

[0154]

[0155] Among them, δx j,t , δy j,t and δz j,t are respectively the response disturbance offsets of the end of the spreader of the j-th tower crane in the X, Y, and Z directions in three-dimensional space after being affected by the shaking of the i-th ship at the t operation time.

[0156] The following are specific examples:

[0157] Assume that the j-th tower crane establishes a corresponding disturbance response coefficient matrix R j As follows, the expression is:

[0158]

[0159] Then

[0160]

[0161] Finally, the offset of the end of the spreader of the j-th tower crane in three-dimensional space corresponding to the i-th ship's predetermined alignment point at the t operation time is

[0162] In this embodiment, by introducing a dynamic disturbance data set, the roll angle, pitch angle, tidal change height, wave interference amplitude and other marine environment disturbance factors of the working ship corresponding to the j-th tower crane at the t working moment are obtained in real time, enabling the onshore tower crane to have the ability to dynamically sense the state of the working target, thus making up for the technical shortcomings of the traditional fixed hoisting system that has no response and compensation to external dynamic disturbances. By identifying the disturbance influence factor unit, the above disturbance factors are mapped to the sensitive response of the offset of the end of the spreader in the X, Y, and Z axis directions. By distinguishing the different response mechanisms of external disturbances of different tower cranes (such as boom length, stiffness) through structural parameters, a personalized disturbance response coefficient matrix is established to realize the personalization of the modeling of the tower crane structure to dynamic target disturbances, the differentiation of responses, and the intelligence of processing. With the help of the linear mapping relationship between the coefficient matrix and the disturbance data set, the system can accurately predict the real-time offset trend of the end of the spreader during the operation process, form a three-dimensional disturbance prediction offset vector, and realize the predictive dynamic correction of the hoisting path. Compared with the traditional method of adjusting by relying on static spreader attitude control or manual experience judgment, this embodiment greatly improves the adaptability and precision control ability of the onshore tower crane to the working target under complex sea conditions without relying on a high-precision ship alignment device. Further, the prediction result can be used as an input for the second avoidance unit and the path dynamic planning module to realize the pre-adjustment of the tower crane operation trajectory, the fine adjustment of the attitude, and the adaptive control of the hoisting speed, avoiding accidents such as hoisting failure, excessive swing of the spreader, or impact on the hull caused by irregular swaying of the ship or tidal level change, and effectively enhancing the stability and safety of the onshore tower crane hoisting operation process.

[0163] Embodiment 7

[0164] This embodiment is an explanatory description carried out in Embodiment 6. Please refer to Figure 1 , specifically, the second avoidance unit is used to target the offset of the end of the spreader of the j-th tower crane in the three-dimensional space at the t working moment relative to the predetermined alignment point of the i-th working ship to perform target correction compensation on the original planned alignment point, specifically:

[0165] Set the original planned alignment point as:

[0166] where and respectively represent the X, Y, and Z axis positions of the j-th tower crane relative to the predetermined alignment point of the i-th working ship;

[0167] For the offset of the end of the spreader of the j-th tower crane in the three-dimensional space at the t working moment relative to the predetermined alignment point of the i-th working ship Compensate the original planned alignment point to obtain the target correction position

[0168]

[0169] Among them, the target correction position This is the actual target position that the tower crane should move to after considering the ship disturbance; it is the new position obtained by adding the ship disturbance compensation to the original planned alignment point;

[0170] And a controller is used to convert the target correction position into the first compensation instruction, including: boom length adjustment amount, trolley displacement adjustment, hoisting height adjustment and speed regulation. Through boom length adjustment, trolley displacement adjustment and hoisting height adjustment, it is ensured that the position of the end of the spreader is accurately adjusted, avoiding hoisting errors caused by ship swaying; considering the continuously changing dynamic disturbance of the ship, the compensation operation has high real-time performance and adaptability, and can make corresponding adjustments according to the real-time operation situation, effectively coping with the complex and changeable sea conditions.

[0171] The first avoidance unit is to avoid the interference risk between tower cranes, and the second avoidance unit, that is: for the end of the spreader of the jth tower crane in three-dimensional space at the t operation moment, the compensation for the offset vector corresponding to the ith ship position during operation, that is: the dynamic compensation mechanism for the alignment error between the tower crane and the ship.

[0172] In this embodiment, the second avoidance unit in this embodiment can effectively adjust the original planned alignment point and optimize the actual alignment target of the end of the tower crane spreader by real-time sensing and dynamic compensation of the offset caused by ship disturbance. In traditional tower crane operations, the dynamic disturbance of the ship often causes deviations in hoisting tasks, resulting in low operation efficiency and even the risk of conflicts. And in this embodiment, by accurately calculating the offset influence of the ship on the end of the tower crane spreader at a specific operation moment, the original planned alignment point is corrected in real time to ensure that the tower crane can continuously perform hoisting tasks at the optimal position. First, based on the sea condition factors such as roll angle, pitch angle, tidal height change and wave interference amplitude in the dynamic disturbance data group, the second avoidance unit calculates the offset of the end of the spreader of the jth tower crane to obtain a new target correction position. Through this dynamic correction, the end of the tower crane spreader can be timely adjusted to an ideal position that conforms to the predetermined operation target, thus avoiding the alignment error caused by ship dynamic interference.

[0173] Embodiment 8

[0174] This embodiment is an explanatory description carried out in Embodiment 7, please refer to Figure 1 , specifically, the inflated voxel correction module is used to obtain the corrected second operation trajectory of the jth tower crane after obtaining the first compensation instruction, and collect and obtain the deviation value between the second operation trajectory and the first operation trajectory to obtain the inflated voxel size △Vp j, used to dilate the corrected second operation trajectory of the j-th tower crane to obtain the second operation trajectory of the j-th tower crane in the floating space coordinate system The dilated second operation trajectory will become a larger area, and the expression is:

[0175]

[0176] When obtaining the second operation trajectory of the j-th tower crane in the floating space coordinate system after that, use the dilated second operation trajectory to perform overlapping voxel detection, and repeat S131 to S132 to reconstruct the interference risk index of the j-th tower crane and the adjacent (j + 1)-th tower crane at the t operation moment And classify it and obtain the corresponding strategy and execute it.

[0177] In this embodiment, the dilation voxel correction module in this embodiment can effectively expand the operation space of the tower crane by performing dilation processing on the corrected operation trajectory, thereby improving operation safety and accuracy. Specifically, the module obtains the second operation trajectory after obtaining the first compensation instruction, calculates the deviation from the original first operation trajectory to determine the dilation voxel size, and then dilates the corrected operation trajectory to obtain a new and safer operation trajectory area. This dilated trajectory shows a larger coverage range in the floating space coordinate system, ensuring that the tower crane can still operate safely in a dynamic environment. By dilating the corrected operation trajectory, it can effectively compensate for the spatial deviation caused by ship dynamic disturbances and the structural errors of the tower crane itself, so as to ensure that the tower crane can perform lifting tasks within a wider space range. The dilated trajectory takes into account potential disturbance factors such as ship movement and climate change, enabling the tower crane operation to be unrestricted by narrow spaces and improving the adaptability of the operation environment. The coverage range of the dilated operation trajectory increases, avoiding interference and collision risks caused by ship dynamic changes or other external factors. By re-performing overlapping voxel detection, it can timely detect potential interference areas between the tower crane and adjacent tower cranes, and classify them according to the latest interference risk index to ensure safety during the operation process and reduce the possibility of accidents.

[0178] It should be noted that: all calculation formulas in this application document adopt, including but not limited to, regression analysis in machine learning algorithms to deeply analyze the relevant parameters collected, identify their natural trends and interrelationships. Using professional software, such as the Scikit-learn library of Python or the R language, automatically generate a mathematical model that matches the data. Then, objectively evaluate the model performance through methods such as cross-validation, and combine continuous feedback and optimization to ensure that the created formula truly reflects the internal laws of the data, thereby ensuring its effectiveness and accuracy, and ensuring that the calculation process conforms to the constraints of natural laws, rather than being based on artificially set rules.

[0179] The technical solution of the present invention can be embodied in the form of a software product in essence or in the part that contributes to the prior art. This computer software product can be stored in a computer-readable storage medium, such as a floppy disk of a computer, a read-only memory (ROM), a random access memory (RAM), a flash memory (FLASH), a hard disk or an optical disc, etc., and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods of various embodiments of the present invention.

[0180] The logic and / or steps represented in the flowchart or described in other ways herein, for example, can be considered as a definite sequence list of executable instructions for implementing logical functions, and can be specifically implemented in any computer-readable medium for use by an instruction execution system, apparatus or device (such as a computer-based system, a system including a processor, or other systems that can fetch and execute instructions from the instruction execution system, apparatus or device), or used in combination with these instruction execution systems, apparatus or devices. For the purposes of this specification, a "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by or in combination with an instruction execution system, apparatus or device.

[0181] It should be noted that the above embodiments are only used to illustrate the technical solution of the present invention and not to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solution of the present invention can be modified or equivalently replaced without departing from the spirit and scope of the technical solution of the present invention, and all of them should be covered by the scope of the claims of the present invention.

[0182] It should be noted that the above embodiments are only used to illustrate the technical solution of the present invention and not to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solution of the present invention can be modified or equivalently replaced without departing from the spirit and scope of the technical solution of the present invention, and all of them should be covered by the scope of the claims of the present invention.

Claims

1. A three-dimensional space collision avoidance system based on multiple tower cranes, characterized in that, Including: A trajectory acquisition module, which is used to establish a three-dimensional space coordinate conversion model for the port operation area, uniformly convert the operation coordinates of each tower crane into a floating space coordinate system with the center of the i-th ship hatch as the reference, and collect and model the first operation trajectory of the j-th tower crane corresponding to the i-th ship in real time, construct a spatial map composed of three-dimensional voxel grids VoxelGrid, and obtain a first horizontal coverage map, a second descending coverage map, a third sector area occupancy map, a fourth track occupancy area map, and a fifth interference area map; A trajectory recognition module is used to perform spatial conflict detection between tower cranes based on a first horizontal coverage map, a second descending coverage map, a third fan-shaped area occupancy map, a fourth track occupancy area map, and a fifth interference area map, and identify an overlapping voxel set with potential spatial interference between the j-th tower crane and the adjacent (j + 1)-th tower crane at the t operation moment. When It indicates that there is a spatial coincidence area at the voxel level at the t operation moment, indicating that there is a potential spatial interference risk between the j-th tower crane and the adjacent (j + 1)-th tower crane at the t operation moment. A first warning instruction is generated, and an interference risk index between the j-th tower crane and the adjacent (j + 1)-th tower crane at the t operation moment is constructed. Then it is graded, and corresponding avoidance strategies are generated according to the grading. A dynamic acquisition module, which collects ship dynamic disturbance information in the port scene and generates a dynamic disturbance data group; A dynamic interference module, which is used to predict and obtain the offset of the end of the spreader of the j-th tower crane in the three-dimensional space relative to the i-th predetermined docking point of the ship corresponding to the operation at the t operation moment based on the dynamic disturbance data group After performing target correction compensation on the original planned docking point, a first compensation instruction is obtained; The dilated voxel correction module is used to obtain the second working trajectory of the j-th tower crane after obtaining the first compensation instruction, collect the deviation value between the second working trajectory and the first working trajectory, and obtain the dilated voxel size △Vp j , which is used to dilate the second working trajectory of the j-th tower crane after correction, and obtain the second working trajectory of the j-th tower crane in the floating space coordinate system The trajectory recognition module repeatedly uses the second working trajectory of the j-th tower crane in the floating space coordinate system after dilation to perform overlapping voxel detection and correction.

2. The three-dimensional space collision avoidance system based on multiple tower cranes according to claim 1, characterized in that: The trajectory acquisition module includes a hull floating reference point acquisition unit; The hull floating reference point acquisition unit is used to obtain the three-dimensional spatial position and attitude angle information of the center of the k-th hatch of the i-th ship, so as to construct a floating space coordinate system based on WebGL, Unity or UE.

3. A three-dimensional space collision avoidance system based on multiple tower cranes according to claim 1, characterized in that: The trajectory acquisition module further includes a three-dimensional voxel network construction unit and a spatial map extraction unit; The three-dimensional voxel network construction unit is used for the scenario where multiple tower cranes operate on the port ship at the same time. It collects the spatial trajectories of the spreaders of each tower crane in the unified floating coordinate system, performs time parameterization modeling, and maps the trajectories of different operation processes to various spatial maps, and performs spatial collision calculations based on three-dimensional voxel grids. The specific steps are as follows: S11. Model the first working trajectory of the spreader of each tower crane in the floating space coordinate system to obtain the first working trajectory of the spreader of the j-th tower crane and the adjacent (j + 1)-th tower crane in the floating space coordinate system and The expression is: where t represents the operation time, and x j (t), y j (t), z j (t) represent the coordinates of the spatial position of the sling of the j-th tower crane in the X, Y, and Z directions at the operation time t; x j+1 (t), y j+1 (t), z j+1 (t) represents the coordinates of the spatial position of the sling of the adjacent (j + 1)-th tower crane in the X, Y, and Z directions at the working moment t. S12. The spatial map extraction unit is used to construct a spatial map, specifically composed of three-dimensional voxel grids VoxelGrid, and specifically includes: S121. Extract the horizontal displacement section of the spreader from the operation trajectory of the spreader of the j-th tower crane in the floating space coordinate system, and map the first operation trajectory of the spreaders of the j-th tower crane and the adjacent (j + 1)-th tower crane in the floating space coordinate system and the corresponding trajectory segments to a three-dimensional voxel grid to obtain the first horizontal coverage map and generate the first horizontal occupancy marker; S122. Extract the path of the spreader of the j-th tower crane descending from the air to the cabin from the trajectory, specifically: z j (t) represents the vertical height of the spreader of the j-th tower crane at the working time t. During the corresponding time period, it is recognized that the spreader is in the lowering operation state, and the recognition rule is: and the spreader speed is not equal to 0; it means that at the working time t, the spreader of the j-th tower crane is moving downward; where dz j (t) represents the change in the vertical position of the spreader of the j-th tower crane at the working time t, that is: z j (t + Δt) - z j (t), dt represents the time interval between two times t and t + Δt; After identifying that the spreader is in the lowering operation state, the lowering trajectory segment is obtained The expression is: Among them, is the start time of the lowering operation, is the end time of the lowering operation; Performing offline equidistant sampling on the lowering trajectory segment to obtain a set of trajectory line segments: Among them, t1, t2,... t n represent n sampling instants, and the time interval of the lower trajectory segment is discrete time points; And generating a set of continuous line segments: Among them, and Every two consecutive sampling points form a lowering line segment, indicating the spatial movement of the spreader during this period; (t k ) and (t k+1 ) represent two consecutive time points, with a fixed time interval between them. (t k ) represents the k-th second, and (t k+1 ) represents the (k + 1)-th second; Segment is a mathematical object describing the linear motion between two points, representing the path of the spreader between two points; For each downward segment L k , use the 3D Bresenham line projection algorithm to map the line segment to a three-dimensional voxel grid, obtain the second downward coverage map, and generate the second downward occupancy marker; S123. Extract the horizontal rotational motion component from the operation trajectory of the spreader of the j-th tower crane in the floating space coordinate system, set each segment of rotational motion as each sector area according to the time frame, map each sector area to the three-dimensional voxel grid, form a third sector area occupancy map, and generate a third rotational occupancy mark; S124. Extract the common track or track intersection area of the operation space from the operation trajectories of the spreaders of the j-th tower crane and the adjacent (j + 1)-th tower crane in the floating space coordinate system, and expand the track width to form a fourth track occupancy area map; S125. Extract the three-dimensional overlapping volume from the operation trajectories of the spreaders of the j-th tower crane and the adjacent (j + 1)-th tower crane in the floating space coordinate system to form a fifth interference area map.

4. A three-dimensional space collision avoidance system based on multiple tower cranes according to claim 3, characterized in that: The trajectory recognition module includes a conflict detection unit and a first avoidance unit; The conflict detection unit is used to count the first horizontal coverage map, the second descending coverage map, the third sector area occupancy map, the fourth track occupancy area map, and the fifth interference area map in S121 - 125, and perform the following operations to realize the spatial conflict detection between tower cranes: S131. Extract the first horizontal coverage map, the second descending coverage map, the third fan-shaped area occupancy map, the fourth track occupancy area map, and the fifth interference area map corresponding to S121 - S125, and identify and obtain the overlapping voxel set with potential spatial interference between the j-th tower crane and the adjacent (j + 1)-th tower crane at the t operation moment The expression is: In the formula, represents the voxel set of the k-th type of occupancy map of the j-th tower crane at the t-th operation moment. When k = 1, it is the first horizontal coverage map; when k = 2, it is the second descending coverage map; when k = 3, it is the third sector area occupancy map; when k = 4, it is the fourth track occupancy area map; when k = 5, it is the fifth interference area map; represents the voxel set of the k-th type of occupancy map of the adjacent (j + 1)-th tower crane at the t-th operation moment; When represents an empty set, indicating that there is a spatial coincidence area at the voxel level at the t operation moment, indicating that there is a potential spatial interference risk between the j-th tower crane and the adjacent (j + 1)-th tower crane at the t operation moment, and generating a first warning instruction; When represents that there is no potential spatial interference risk between the j-th tower crane and the adjacent (j + 1)-th tower crane, and there is no spatial coincidence in the current operation trajectory at the t operation moment; S132. After receiving the first warning instruction, extract the set of overlapping voxels where there is a potential spatial interference between the j-th tower crane and the adjacent (j + 1)-th tower crane at the t operation moment the number of voxels in the overlapping area in, and construct the interference risk index of the j-th tower crane and the adjacent (j + 1)-th tower crane at the t operation moment through the following formula where |V safe | represents the set maximum number of voxel in the spatial redundancy safety region, and the result of GI j,j+1 obtains a normalized value in the range of [0, 1], which is convenient for grading.

5. A three-dimensional space collision avoidance system based on multiple tower cranes according to claim 4, characterized in that: The first avoidance unit is used to classify the value of the interference risk index of the j-th tower crane and the adjacent (j + 1)-th tower crane at the t-th operation moment, including: values, including: When it indicates that there is no intersection in the spatial operation voxels of the two tower cranes at the t operation moment, generating the first safety level, indicating that the j-th tower crane and the adjacent (j + 1)-th tower crane continue to operate normally; When it indicates that there is an overlap in the edge area between the two tower cranes at the t operation moment, and a second low-risk level strategy is generated, including: recording the status and continuing the operation, reducing the boom amplitude of the j-th tower crane by 3%-5%, reducing the trolley position hook speed by 3%-5%, adjusting the lifting height by 3%-5% in the opposite direction of the overlap, and smoothly reducing the overall speed by 10%-20%; When occurs, the interference area enters the actual operation path of the tower arm / lifting appliance, has the potential for physical interference, and is marked as a conflict area, generating a third-level risk strategy, including: generating a second warning instruction, offsetting 6%-10% in the opposite direction of the overlap, reducing the boom amplitude of the jth tower crane by 10%-20%, lowering the hook speed of the trolley position by 6%-10%, adjusting the lifting height by 6%-10% in the opposite direction of the overlap, and smoothly reducing the overall speed by 21%-30%; when When , it means that there is a risk of collision between the two tower cranes at the operation time t, and it is marked as a conflict area, generating the fourth highest risk level strategy, including: generating the third warning instruction, adjusting the tower crane near the edge of the conflict area to give priority to avoidance, adjusting the arm of the tower crane to retract ≥ 25% of the arm extension, immediately withdrawing from the conflict area, adjusting the lifting height in the opposite direction of the overlap by 30%, leaving the dangerous intersection, and smoothly reducing the speed by 40%-50%; and the tower crane near the central axis will synchronously suspend the lifting operation; and then judge the interference risk index again. Until After the interference is lifted, the jth tower crane resumes operation through the intervention of the master control.

6. The three-dimensional space collision avoidance system based on multiple tower cranes according to claim 1, characterized in that: The dynamic disturbance data group includes: the roll angle of the i-th ship corresponding to the spreader of the j-th tower crane at the t operation moment the pitch angle θ i,t , the tidal change height Δh tide,t and the wave interference amplitude A wave, t .

7. A three-dimensional space collision avoidance system based on multiple tower cranes according to claim 6, characterized in that:

7. The dynamic interference module includes a disturbance influence factor identification unit and a second avoidance unit; The recognition of disturbance influence factor unit is used to extract the roll angle of the i-th ship corresponding to the spreader of the j-th tower crane at the t operation moment in the dynamic disturbance data group Pitch angle θ i,t , tidal change height Δh tide,t and wave interference amplitude A wave,t . Since different tower cranes have different structures, different arm lengths, different stiffnesses, and different disturbance effects. As a shore-based fixed device, the structure of the tower crane will not "move by itself" due to the shaking of the ship, tides or waves. However, since the object of the tower crane's lifting operation, that is, the ship, is in dynamic motion, it is necessary to "perceive these disturbances" and thus make intelligent response operations. It is necessary to establish a corresponding disturbance response coefficient matrix R for the j-th tower crane j , and the expression is: where r 11 , r 12 , r 13 , r 14 are respectively the response sensitive offset values of the roll angle change to the X-direction offset, the response sensitive offset value of the pitch angle to the X-direction offset, the response sensitive offset value of the tidal change height to the X-direction offset, and the response sensitive offset value of the wave interference amplitude to the X-direction offset; where r 21 , r 22 , r 23 , r 24 are respectively the sensitive offset value of the roll angle change to the Y-direction offset, the sensitive offset value of the pitch angle to the Y-direction offset, the sensitive offset value of the tidal change height to the Y-direction offset, and the sensitive offset value of the wave interference amplitude to the Y-direction offset; where r 31 , r 32 , r 33 , r 34 are respectively the sensitive offset value of the roll angle change to the Z - direction offset, the sensitive offset value of the pitch angle to the Z - direction offset, the sensitive offset value of the tidal change height to the Z - direction offset, and the sensitive offset value of the wave interference amplitude to the Z - direction offset; And establish the corresponding disturbance response coefficient matrix R according to the j-th tower crane j Convert it into vector form and predict the offset of the j-th tower crane hoist end in three-dimensional space at the time of operation t, corresponding to the scheduled alignment point of the i-th ship. The expression is: Among them, δx j,t , δy j,t and δz j,t are the response disturbance offsets of the end of the spreader of the j-th tower crane in the X, Y, and Z directions in three-dimensional space after being affected by the shaking of the i-th ship at the t operation moment, respectively.

8. A three-dimensional space collision avoidance system based on multiple tower cranes according to claim 7, characterized in that: The second avoidance unit is used for the offset of the end of the spreader of the j-th tower crane in the three-dimensional space at the t-th operation moment relative to the i-th predetermined ship docking point corresponding to the operation Perform target correction compensation on the originally planned docking point, specifically: The originally planned target site is set as: Among them, and respectively represent the X, Y, and Z axis positions of the j-th tower corresponding to the i-th predetermined docking point of the ship. The offset of the end of the spreader of the j-th tower crane in the three-dimensional space at the t-th operation moment relative to the i-th ship's predetermined docking point corresponding to the operation Compensate for the original planned docking point to obtain the target corrected position Among them, the target correction position This is the actual target position that the tower crane should move to after considering the ship disturbance; it is the new position obtained by adding the ship disturbance compensation to the original planned alignment point; And a controller is used to convert the target correction position into a first compensation instruction, including: boom length adjustment amount, trolley displacement adjustment, lifting height adjustment, and speed adjustment.

9. The three-dimensional space collision avoidance system based on multiple tower cranes according to claim 1, characterized in that: The inflation voxel correction module is used to obtain the second operation trajectory of the j-th tower crane after correction after obtaining the first compensation instruction, collect and obtain the deviation value between the second operation trajectory and the first operation trajectory, and obtain the inflation voxel size △Vp j , and is used to inflate the second operation trajectory of the j-th tower crane after correction to obtain the second operation trajectory of the j-th tower crane in the floating space coordinate system The inflated second operation trajectory will become a larger area, and the expression is: After obtaining the second operation trajectory of the j-th tower crane in the floating space coordinate system After that, use the expanded second operation trajectory to perform overlapping voxel detection, and repeat S131 to S132 to reconstruct the interference risk index of the j-th tower crane and the adjacent (j + 1)-th tower crane at the t operation moment And classify it, obtain the corresponding strategy and execute it.

Images