A collaborative sensing and autonomous operation method for material in the auger unloader based on a multi-source laser scanning array

Abstract

This invention relates to a collaborative sensing and autonomous operation method for cargo hold material in a spiral unloader based on a multi-source laser scanning array. It solves the problems of large blind spots, high positioning errors, and poor hatch recognition stability in existing single-laser scanning systems. It includes: S1, heterogeneous laser scanning array networking; S2, pose differential IMU deep coupling for motion point cloud distortion suppression; S3, manual target point binding; S4, point cloud data processing; and S5, multi-level collision avoidance control. The advantages of this invention are: significantly improved hatch recognition stability and coal pile model construction accuracy; point cloud distortion prevention in dynamic environments, effectively reducing manual intervention and accident rates; meeting the technical requirements of unmanned unloading operations; and strong industrial adaptability.

Description

Technical Field

[0001] This invention relates to the field of automated loading and unloading technology, specifically to a method for collaborative sensing and autonomous operation of cargo in a spiral unloader based on a multi-source laser scanning array. Background Technology

[0002] Currently, bulk cargo unloading operations at ports mainly rely on manual labor, resulting in low efficiency and high safety risks. Although laser scanning technology is gradually being applied to automated unloading, in the operation of spiral continuous unloaders, due to the complex structure of the ship's hold, irregular material accumulation, and severe environmental interference, existing single laser scanning systems generally suffer from problems such as large scanning blind spots, high positioning errors, and poor hatch recognition stability, making it difficult to meet the requirements for unmanned and high-efficiency operations. Summary of the Invention

[0003] The purpose of this invention is to address the problems existing in the prior art by proposing a collaborative sensing and autonomous operation method for cargo hold of a screw unloader based on a multi-source laser scanning array. The method employs a multi-source laser scanning array collaborative sensing method combined with a manual target marking mechanism and a dynamic distortion compensation algorithm to achieve high-precision modeling and autonomous operation control of the screw unloader across the entire cargo hold under complex working conditions, thereby improving unloading efficiency and operational safety.

[0004] To achieve the above objectives, the present invention adopts the following technical solution: a method for collaborative sensing and autonomous operation of cargo in a spiral unloader based on a multi-source laser scanning array, comprising the following steps:

[0005] S1, heterogeneous laser scanning array networking;

[0006] S2, pose difference-IMU deep coupling for motion point cloud distortion suppression;

[0007] S3, Artificially labeled target binding;

[0008] S4, Point Cloud Data Processing;

[0009] S5, multi-level collision avoidance control.

[0010] In step S1, the heterogeneous laser scanning array network includes a 3D laser scanner layout, a 2D tactile laser scanner layout, and blind spot safety monitoring coverage.

[0011] In the aforementioned method for collaborative sensing and autonomous operation of material in a spiral unloader's hold based on a multi-source laser scanning array, the 3D laser scanner layout includes:

[0012] S011. Four 3D scanners are symmetrically arranged at the four corners of the rotating platform, forming a trapezoidal layout.

[0013] S012, each device has a field of view of 120°×25°, and the overall network achieves full coverage of every hatch corner.

[0014] S013. Use the dock coordinate system to calibrate the external parameters of each scanner to achieve real-time fusion of multi-source point clouds in a unified space.

[0015] In the aforementioned collaborative sensing and autonomous operation method for material in a spiral unloader's hold based on a multi-source laser scanning array, the 2D stage laser scanner layout includes:

[0016] The S021 2D pendulum laser scanner is mounted at the end of a horizontal arm and is driven to swing via a servo controller, with an angle range of ±90°.

[0017] The S022 2D staged laser scanner has a scanning surface perpendicular to the longitudinal axis of the ship's hull, used to penetrate blind spots in the coal pile for blind spot scanning.

[0018] S023. Set the swing frequency to 2Hz to meet the contour acquisition requirements of the bilge and corner areas.

[0019] The coverage of blind spot security monitoring includes:

[0020] S031. By linking the scheduling of the platform angle with the unloading route planning, intermittent coverage of blind spots can be achieved;

[0021] S032. Dynamic triangular monitoring zones are formed in high-risk areas such as unloading heads, bulkheads, and crossbeams.

[0022] In the above-mentioned collaborative sensing and autonomous operation method for material in the auger unloader based on a multi-source laser scanning array, step S2 includes the following steps:

[0023] S21. Synchronous acquisition of multi-source data: The 3D laser scanner and the 2D staged laser scanner simultaneously acquire IMU attitude information and GNSS differential positioning data, with data synchronization accuracy controlled within 10ms.

[0024] S22. Pose Difference Calculation: Using the laser scanner mounting point as a reference, calculate its pose changes Δθ, Δx, Δy, and Δz across consecutive time frames, simultaneously generating a time-difference pose transformation matrix for post-processing compensation of positional drift during scanning.

[0025] S23. Deeply Coupled Compensation Algorithm: This algorithm embeds IMU attitude changes into a point cloud projection algorithm in real time, correcting the spatial mapping results of each frame's point cloud. Kalman filtering is then used to fuse and filter the GNSS-IMU data, further reducing interference and jitter.

[0026] S24. Effect evaluation: The deep coupling compensation algorithm reduces the maximum distortion error of the swing arm scanning point cloud from more than 200mm to less than 100mm.

[0027] In the above-mentioned collaborative sensing and autonomous operation method for material in the auger unloader based on a multi-source laser scanning array, step S3 includes the following steps:

[0028] S31. Target Deployment: High-reflectivity active targets with a reflectivity ≥90% are deployed at the ends of the reinforcing ribs of the barge hatches and in structurally stable areas.

[0029] S32. Initial Modeling and Binding: Before the first entry into the hatch, scan the target location and use the point cloud intensity thresholding method to extract high-reflectivity areas.

[0030] S33. Calculate the three-dimensional coordinates of its centroid as the target reference point, and perform rigid coordinate registration with the hatch boundary to establish the spatial transformation matrix T.

[0031] S334, Dynamic Recognition and Coordinate Mapping: Each time the barge is entered for operation, the target position is scanned and identified first. By calculating the deviation from the initial T matrix, the unloader's operating space is automatically and in real time corrected.

[0032] In the above-mentioned collaborative sensing and autonomous operation method for material in the auger unloader based on a multi-source laser scanning array, step S4 includes the following steps:

[0033] S41, Dynamic Filtering;

[0034] S42, Multi-source point cloud registration;

[0035] S43, Feature Extraction;

[0036] S44, Model Reconstruction.

[0037] In the above-mentioned collaborative sensing and autonomous operation method for material in the auger unloader based on a multi-source laser scanning array, step S41 includes the following steps:

[0038] S411. Perform statistical outlier removal on the collected raw point cloud data.

[0039] S412. Calculate the Euclidean distance between each point and its k nearest neighbors. If the distance deviates from the mean by more than a set threshold, it is considered noise.

[0040] S413, Typical parameters: k=50, standard deviation multiple threshold set to 1.0-2.0.

[0041] S414. The proportion of effective point cloud retained after processing is increased to over 95%;

[0042] Step S42 includes the following steps:

[0043] S421. Use the point cloud of a 3D laser scanner at a fixed position as the global reference frame.

[0044] S422. Use an improved iterative nearest-point algorithm to perform coordinate transformation on the local point cloud generated by the 2D staged laser scanner.

[0045] S423. The matching strategy uses edge features and curvature to jointly control error convergence.

[0046] S424, The final registration error is controlled within 0.5 degrees of angular accuracy and within 30 mm of linear accuracy;

[0047] In the above-mentioned collaborative sensing and autonomous operation method for material in the auger unloader based on a multi-source laser scanning array, step S43 includes the following steps:

[0048] S431. Hatch edge recognition uses a straight line detection algorithm based on Hough transform.

[0049] S432. A region growing clustering method is used for the coal pile area, and point cloud intensity and geometric curvature are introduced as dual features for joint extraction.

[0050] S433, Intensity reflects the reflective target or metal boundary, and geometric information is used to distinguish between the coal pile and the cabin structure;

[0051] Step S44 includes the following steps:

[0052] S441. The reconstruction results are presented in the form of a three-dimensional raster, constructing a digital twin model of the barge compartment.

[0053] S442. Use the Octree voxel filtering method to downsample and optimize the model, thereby improving modeling efficiency.

[0054] S443, with an update frequency of 6 times per second, meets the real-time requirements for unloading path planning and collision avoidance control.

[0055] In the above-mentioned collaborative sensing and autonomous operation method for material in the auger unloader based on a multi-source laser scanning array, step S5 includes the following steps:

[0056] S51. Real-time monitoring of the relative distance between the unloading head and key parts of the bulkhead and crossbeams;

[0057] S52, exceeding the limit triggers emergency deceleration or emergency stop, with a response time of less than 0.3 seconds.

[0058] Compared with existing technologies, the advantages of this invention are: significantly improved hatch identification stability and coal pile model construction accuracy; point cloud no longer distorts in dynamic environments, and system positioning stability reaches 99.2%; operation efficiency is improved by about 22%, effectively reducing manual intervention and accident rate; it meets the technical requirements of unmanned unloading operations and has strong industrial adaptability. Attached Figure Description

[0059] Figure 1 This is a flowchart of the method of the present invention;

[0060] Figure 2 This is a schematic diagram of the hardware deployment in this invention;

[0061] Figure 3 This is a schematic diagram of the deployment of the 3D laser scanner in this invention;

[0062] Figure 4 This is a schematic diagram of the deployment of artificially marked target points in this invention;

[0063] Figure 5 This is a schematic diagram of the deployment of the 2D laser scanner in this invention;

[0064] Figure 6 This is a scanned point cloud image of the barge in this invention;

[0065] Figure 7 This is a point cloud image scanned by the 2D laser scanner in this invention. Detailed Implementation

[0066] The present invention will now be described in further detail with reference to the accompanying drawings and specific embodiments.

[0067] like Figure 1-2 As shown, a collaborative sensing and autonomous operation method for material in the auger unloader's hold based on a multi-source laser scanning array includes the following steps:

[0068] S1, heterogeneous laser scanning array networking;

[0069] S2, pose difference-IMU deep coupling for motion point cloud distortion suppression;

[0070] S3, Artificially labeled target binding;

[0071] S4, Point Cloud Data Processing;

[0072] S5, multi-level collision avoidance control.

[0073] like Figure 4 As shown, artificially marked target points use high-reflectivity markers to improve scanning recognition quality and efficiency.

[0074] In step S1, the heterogeneous laser scanning array network includes a 3D laser scanner layout, a 2D tactile laser scanner layout, and blind spot safety monitoring coverage.

[0075] like Figure 3 As shown, the 3D laser scanner layout includes:

[0076] S011. Four 3D scanners are symmetrically arranged at the four corners of the rotating platform, forming a trapezoidal layout.

[0077] S012, each device has a field of view of 120°×25°, and the overall network achieves full coverage of every hatch corner.

[0078] S013. Use the dock coordinate system to calibrate the external parameters of each scanner to achieve real-time fusion of multi-source point clouds in a unified space;

[0079] like Figure 5 As shown, the layout of the 2D staged laser scanner includes:

[0080] The S021 2D pendulum laser scanner is mounted at the end of a horizontal arm and is driven to swing via a servo controller, with an angle range of ±90°.

[0081] The S022 2D staged laser scanner has a scanning surface perpendicular to the longitudinal axis of the ship's hull, used to penetrate blind spots in the coal pile for blind spot scanning.

[0082] S023. Set the swing frequency to 2Hz to meet the contour acquisition requirements of the bilge and corner areas.

[0083] The coverage of blind spot security monitoring includes:

[0084] S031. By linking the scheduling of the platform angle with the unloading route planning, intermittent coverage of blind spots can be achieved;

[0085] S032. Dynamic triangular monitoring zones are formed in high-risk areas such as the unloading head, bulkhead, and crossbeams. This achieves a sensing coverage rate of no less than 98%, effectively addressing high-obstruction operating conditions.

[0086] To achieve blind-spot-free perception of the entire barge hold by the spiral unloader, this invention uses multiple types of 3D laser scanners and 2D swivel laser scanners with different field of view ranges and deployment angles, distributed in multiple functional parts of the equipment according to the unloader's preset operating logic, thereby forming a complementary, multi-angle, high-density three-dimensional perception network for the complex barge hold space.

[0087] The system integrates multiple 3D laser scanners with a wide horizontal field of view (120°×25°) for full cabin contour modeling and positioning benchmark establishment, as well as a 2D laser scanner with a servo stage, forming a laser scanning network solution of "heterogeneous acquisition, heterogeneous blind spot filling and heterogeneous fusion".

[0088] Step S2 includes the following steps:

[0089] S21. Synchronous acquisition of multi-source data: The 3D laser scanner and the 2D oscillating laser scanner synchronously acquire IMU attitude information (roll angle, pitch angle, yaw angle) and GNSS differential positioning data, with data synchronization accuracy controlled within 10ms.

[0090] S22. Pose difference calculation: Taking the laser scanner mounting point as the reference, calculate its pose changes Δθ, Δx, Δy, and Δz in continuous time frames, and form a time difference pose transformation matrix for post-processing to compensate for the position drift at the scanning point.

[0091] The laser scanners here include 3D laser scanners and 2D tabletop laser scanners.

[0092] S23. Deep Coupling Compensation Algorithm: The IMU attitude change is embedded in the point cloud projection algorithm in real time to correct the spatial mapping result of the point cloud in each frame. Then, Kalman filtering is used to fuse and filter the GNSS-IMU data to further reduce interference jitter.

[0093] S24. Effect evaluation: The deep coupling compensation algorithm reduces the maximum distortion error of the swing arm scanning point cloud from more than 200mm to less than 100mm;

[0094] It meets the accuracy requirements of the anti-collision control system and improves the accuracy of unloading path planning.

[0095] To address the distortion of 2D laser scanning point clouds caused by horizontal arm swing, a compensation algorithm based on pose difference and deep coupling of IMU is proposed. GNSS RTK and IMU are used to perform attitude compensation for horizontal arm swing and low-frequency vibration, thereby achieving dynamic point cloud coordinate stabilization.

[0096] Step S3 includes the following steps:

[0097] S31. Target deployment: High reflectivity active targets with a reflectivity ≥90% are deployed at the ends of the reinforcing ribs at the barge hatch and in the structurally stable area. The targets are square in shape, with a length and width of approximately 500mm, and the surface is coated with a mirror-grade film material for easy laser identification.

[0098] like Figure 6-7 As shown, point cloud data of the barge hatch area is collected using a heterogeneous laser scanning array network, and the high-intensity point cloud area corresponding to the target is extracted by the intensity threshold segmentation method.

[0099] S32. Initial Modeling and Binding: Before the first entry into the hatch, scan the target location and use the point cloud intensity thresholding method to extract the high reflectivity area;

[0100] S33. Calculate the three-dimensional coordinates of its centroid as the target reference point, and perform rigid coordinate registration with the hatch boundary to establish a spatial transformation matrix T.

[0101] In step S33, the centroid coordinates of the target point cloud are calculated and used as a reference point for hatch space calibration. During the initial calibration process, the spatial rigidity transformation relationship between the target and the edge of the barge hatch is obtained, thereby establishing the spatial mapping matrix T.

[0102] S334, Dynamic Recognition and Coordinate Mapping: Each time the barge is entered for operation, the target position is scanned and identified first. By calculating the deviation from the initial T matrix, the unloader's operating space is automatically and in real time corrected.

[0103] That is, in subsequent unloading operations, by re-identifying the target position and dynamically deriving the spatial position and attitude of the barge hatch in the dock coordinate system based on the transformation matrix T, stable positioning of the barge hull in a changing environment can be achieved.

[0104] This mechanism improves hatch positioning stability from 85% to 99%, significantly suppressing the disturbance effect of hull rolling on the working coordinates.

[0105] High reflectivity targets are placed at the reinforcing ribs of the barge hatch. The rigid spatial relationship between the dock and the hatch is established by identifying the three-dimensional centroid coordinates of the targets. The manual marker binding mechanism is used to solve the coordinate drift problem caused by hull drift, scan interruption or structural obstruction.

[0106] Step S4 includes the following steps:

[0107] S41, Dynamic Filtering;

[0108] S42, Multi-source point cloud registration;

[0109] S43, Feature Extraction;

[0110] S44, Model Reconstruction.

[0111] Step S41 includes the following steps:

[0112] S411. Perform statistical outlier removal on the collected raw point cloud data.

[0113] S412. Calculate the Euclidean distance between each point and its k nearest neighbors. If the distance deviates from the mean by more than a set threshold, it is considered noise (such as floating points caused by dust or fog).

[0114] S413, Typical parameters: k=50, standard deviation multiple threshold set to 1.0-2.0.

[0115] S414. The proportion of effective point cloud retained after processing is increased to over 95%;

[0116] Step S42 includes the following steps:

[0117] S421. Use the point cloud of a 3D laser scanner at a fixed position as the global reference frame.

[0118] S422. Use the improved Iterative Closest Point (ICP) algorithm to perform coordinate transformation on the local point cloud generated by the 2D staged laser scanner.

[0119] S423. The matching strategy uses edge features and curvature to jointly control error convergence.

[0120] S424, The final registration error is controlled within 0.5 degrees of angular accuracy and within 30 mm of linear accuracy;

[0121] Step S43 includes the following steps:

[0122] S431. Hatch edge recognition uses a straight line detection algorithm based on Hough transform.

[0123] S432. A region growing clustering method is used for the coal pile area, and point cloud intensity and geometric curvature are introduced as dual features for joint extraction.

[0124] S433, Intensity reflects the reflective target or metal boundary, and geometric information is used to distinguish between the coal pile and the cabin structure;

[0125] Step S44 includes the following steps:

[0126] S441. The reconstruction results are presented in the form of a three-dimensional raster, constructing a digital twin model of the barge compartment.

[0127] S442. Use the Octree voxel filtering method to downsample and optimize the model, thereby improving modeling efficiency.

[0128] S443, with an update frequency of 6 times per second, meets the real-time requirements for unloading path planning and collision avoidance control.

[0129] Currently, through field testing, the hatch recognition rate has been improved to 98.7%, and the modeling error is controlled within ±2.8% (<100mm).

[0130] Step S5 includes the following steps:

[0131] S51. Real-time monitoring of the relative distance between the unloading head and key parts of the bulkhead and crossbeams;

[0132] S52, exceeding the limit triggers emergency deceleration or emergency stop, with a response time of less than 0.3 seconds.

[0133] In summary, the principle of this embodiment is as follows: Multiple 3D laser scanners are deployed on the rotating platform to achieve full hatch scanning coverage, and a 2D laser scanner is deployed at the end of the horizontal arm, with occlusion compensation achieved through a swing mechanism. GNSSRTK and IMU are combined to achieve dynamic attitude compensation of the point cloud coordinates. Furthermore, a rigid mapping between the barge hatch and the dock coordinate system is constructed by manually setting reflective targets. The collected point cloud is then filtered, multi-source registered, feature extracted, and digitally modeled to drive path generation and collision avoidance control for the unloading operation. By integrating multiple 3D and 2D laser scans, deep coupling of differential positioning and IMU, a manual target binding mechanism, and multi-level collision avoidance control logic, a full-hull-domain digital twin is constructed, significantly improving the intelligence level of coal pile identification, hatch modeling, and autonomous unloading, demonstrating high potential for industrial applications.

[0134] The specific embodiments described herein are merely illustrative of the spirit of the invention. Those skilled in the art to which this invention pertains may make various modifications or additions to the described specific embodiments or use similar methods to substitute them, without departing from the spirit of the invention or exceeding the scope defined by the appended claims.

Claims

1. A method for collaborative sensing and autonomous operation of material in a screw unloader's hold based on a multi-source laser scanning array, characterized in that, Includes the following steps: S1, heterogeneous laser scanning array networking; S2, pose difference-IMU deep coupling for motion point cloud distortion suppression; S3, Artificially labeled target binding; S4, Point Cloud Data Processing; S5, multi-level collision avoidance control; In step S1, the heterogeneous laser scanning array network includes a 3D laser scanner layout, a 2D staged laser scanner layout, and blind spot security monitoring coverage. The layout of the 3D laser scanner includes: S011. Four 3D scanners are symmetrically arranged at the four corners of the rotating platform, forming a trapezoidal layout. S012, each device has a field of view of 120°×25°, and the overall network achieves full coverage of every hatch corner. S013. Use the dock coordinate system to calibrate the external parameters of each scanner to achieve real-time fusion of multi-source point clouds in a unified space; The layout of the 2D staged laser scanner includes: The S021 2D pendulum laser scanner is mounted at the end of a horizontal arm and is driven to swing via a servo controller, with an angle range of ±90°. The S022 2D staged laser scanner has a scanning surface perpendicular to the longitudinal axis of the ship's hull, used to penetrate blind spots in the coal pile for blind spot scanning. S023. Set the swing frequency to 2Hz to meet the contour acquisition requirements of the bilge and corner areas. The aforementioned blind spot security monitoring coverage includes: S031. By linking the scheduling of the platform angle with the unloading route planning, intermittent coverage of blind spots can be achieved; S032. Form dynamic triangular monitoring zones in high-risk areas such as unloading heads, bulkheads, and crossbeams; Step S2 includes the following steps: S21. Synchronous acquisition of multi-source data: The 3D laser scanner and the 2D staged laser scanner simultaneously acquire IMU attitude information and GNSS differential positioning data, with data synchronization accuracy controlled within 10ms. S22. Pose Difference Calculation: Using the laser scanner mounting point as a reference, calculate its pose changes Δθ, Δx, Δy, and Δz across consecutive time frames, simultaneously generating a time-difference pose transformation matrix for post-processing compensation of positional drift during scanning. S23. Deeply Coupled Compensation Algorithm: This algorithm embeds IMU attitude changes into a point cloud projection algorithm in real time, correcting the spatial mapping results of each frame's point cloud. Kalman filtering is then used to fuse and filter the GNSS-IMU data, further reducing interference and jitter. S24. Effect evaluation: The deep coupling compensation algorithm reduces the maximum distortion error of the swing arm scanning point cloud from more than 200mm to less than 100mm.

2. The method for collaborative sensing and autonomous operation of material in a spiral unloader's hold based on a multi-source laser scanning array, as described in claim 1, is characterized in that... Step S3 includes the following steps: S31. Target Deployment: High-reflectivity active targets with a reflectivity ≥90% are deployed at the ends of the reinforcing ribs of the barge hatches and in structurally stable areas. S32. Initial Modeling and Binding: Before the first entry into the hatch, scan the target location and use the point cloud intensity thresholding method to extract high-reflectivity areas. S33. Calculate the three-dimensional coordinates of its centroid as the target reference point, and perform rigid coordinate registration with the hatch boundary to establish the spatial transformation matrix T. S334, Dynamic Recognition and Coordinate Mapping: Each time the barge is entered for operation, the target position is scanned and identified first. By calculating the deviation from the initial T matrix, the unloader's operating space is automatically and in real time corrected.

3. The method for collaborative sensing and autonomous operation of material in a spiral unloader's hold based on a multi-source laser scanning array, as described in claim 1, is characterized in that... Step S4 includes the following steps: S41, Dynamic Filtering; S42, Multi-source point cloud registration; S43, Feature Extraction; S44, Model Reconstruction.

4. The method for collaborative sensing and autonomous operation of material in a spiral unloader's hold based on a multi-source laser scanning array, as described in claim 3, is characterized in that... Step S41 includes the following steps: S411. Perform statistical outlier removal on the collected raw point cloud data. S412. Calculate the Euclidean distance between each point and its k nearest neighbors. If the distance deviates from the mean by more than a set threshold, it is considered noise. S413, Typical parameters: k=50, standard deviation multiple threshold set to 1.0-2.

0. S414. The proportion of effective point cloud retained after processing is increased to over 95%; Step S42 includes the following steps: S421. Use the point cloud of a 3D laser scanner at a fixed position as the global reference frame. S422. Use an improved iterative nearest-point algorithm to perform coordinate transformation on the local point cloud generated by the 2D staged laser scanner. S423. The matching strategy uses edge features and curvature to jointly control error convergence. S424, the final registration error is controlled within 0.5 degrees of angular accuracy and within 30 mm of linear accuracy.

5. The method for collaborative sensing and autonomous operation of material in a spiral unloader's hold based on a multi-source laser scanning array, as described in claim 4, is characterized in that... Step S43 includes the following steps: S431. Hatch edge recognition uses a straight line detection algorithm based on Hough transform. S432. A region growing clustering method is used for the coal pile area, and point cloud intensity and geometric curvature are introduced as dual features for joint extraction. S433, Intensity reflects the reflective target or metal boundary, and geometric information is used to distinguish between the coal pile and the cabin structure; Step S44 includes the following steps: S441. The reconstruction results are presented in the form of a three-dimensional raster, constructing a digital twin model of the barge compartment. S442. Use the Octree voxel filtering method to downsample and optimize the model, thereby improving modeling efficiency. S443, with an update frequency of 6 times per second, meets the real-time requirements for unloading path planning and collision avoidance control.

6. The method for collaborative sensing and autonomous operation of material in a spiral unloader's hold based on a multi-source laser scanning array, as described in claim 1, is characterized in that... Step S5 includes the following steps: S51. Real-time monitoring of the relative distance between the unloading head and key parts of the bulkhead and crossbeams; S52, exceeding the limit triggers emergency deceleration or emergency stop, with a response time of less than 0.3 seconds.

Images