A conveying line article dynamic positioning method and system based on multi-modal ranging

Abstract

The application provides a conveying line article dynamic positioning method and system based on multi-modal ranging, comprising: in response to the start of the conveying line, detecting the to-be-conveyed articles on the conveying line based on a multi-source detection unit to obtain multiple to-be-conveyed article data; performing multi-source fusion processing on the multiple to-be-conveyed article data according to a preset weight to obtain to-be-conveyed article position fusion data; determining a to-be-conveyed article positioning direction based on the to-be-conveyed article position fusion data and a preset positioning position; and driving a power adjustment unit to push the to-be-conveyed article to the preset positioning position according to the to-be-conveyed article positioning direction. The application can improve positioning error, reduce misjudgment rate, realize compatible recognition of multiple specifications and multiple material to-be-conveyed articles, improve positioning success rate, improve detection accuracy, and form a redundancy mechanism through multi-sensing fusion, thereby greatly improving system practicability.

Description

Technical Field

[0001] This invention belongs to the field of transmission line technology, specifically relating to a method and system for dynamic positioning of items on a conveyor line based on multimodal ranging. Background Technology

[0002] In the field of logistics automation positioning technology, especially in the context of goods conveyor lines, accurate positioning of items to be transported is a crucial link in ensuring the efficient operation of loading and unloading, warehousing, and AGV transfer. However, the items to be transported are diverse in specifications, covering various sizes and complex shapes, including special types such as metal items and items with pallets. They also face challenges from complex environments such as dust, water mist, low temperatures, transparent goods covering, and low light conditions, which places extremely high demands on positioning technology.

[0003] Existing positioning technologies have many limitations, often lacking reliability. Their performance also falls short of practical requirements when faced with obstructed transported items or low point cloud accuracy. Furthermore, existing item positioning systems lack flexibility when dealing with items of different sizes, and some devices are prone to jamming or excessive compression during the positioning process, severely impacting production efficiency.

[0004] In summary, existing technologies are insufficient to meet the positioning requirements of "multi-specification, multi-environment, and high reliability" for items, and there is an urgent need for a dynamic positioning solution for multi-specification items that can adapt to complex scenarios. Summary of the Invention

[0005] To alleviate the above problems, the main objective of this application is to propose a method and system for dynamic positioning of items on a conveyor line based on multimodal ranging. The positioning method includes:

[0006] In response to the start of the conveyor line, the multi-source detection unit detects the items to be conveyed on the conveyor line to obtain various data on the items to be conveyed;

[0007] The data of the various items to be transported are fused from multiple sources according to preset weights to achieve compatible identification of items of various specifications and materials, so as to obtain the location fusion data of the items to be transported.

[0008] Based on the fused location data of the item to be transported and the preset positioning location, the positioning direction of the item to be transported is determined;

[0009] According to the positioning direction of the item to be transported, the drive power adjustment unit pushes the item to be transported to the preset positioning position.

[0010] Optionally, the multi-source detection unit includes a laser sensor; the various data on the items to be transported includes the size, edge coordinates, and tilt angle of the items to be transported; the step of detecting the items to be transported on the conveyor line based on the multi-source detection unit in response to the start of the conveyor line, and prior to this step, includes:

[0011] The laser sensor is driven to emit a continuous laser beam, which scans the edge of the item to be transported in real time, causing the edge of the item to block the laser and form diffuse reflection.

[0012] The system receives reflected signals, calculates the distance to the item to be transported using triangulation or time-of-flight methods, and outputs the size, edge coordinates, and tilt angle of the item to be transported.

[0013] Optionally, the multi-source detection unit includes a high-definition camera; the various data on the items to be transported include the geometric center of the items to be transported, the distance from the edge contour of the items to be transported to the centerline of the conveyor line or a preset position; the step of detecting the items to be transported on the conveyor line based on the multi-source detection unit in response to the start of the conveyor line to obtain various data on the items to be transported, and the steps prior to this include:

[0014] A multi-view high-definition visual inspection system is formed by using multiple high-definition cameras at preset shooting heights, distributed at different positions and tilted at preset shooting angles to shoot the top of the items to be transported.

[0015] The edge contour of the item to be transported is extracted by an edge detection algorithm, and the geometric center of the item to be transported and the distance from the edge contour of the item to the centerline of the transport line or the preset position are calculated by combining Hough transform.

[0016] Optionally, the multi-source detection unit includes a microwave sensor; the various data on the items to be transported includes real-time dimensions and positioning data of the items to be transported; the step of detecting the items to be transported on the conveyor line based on the multi-source detection unit in response to the start of the conveyor line, and prior to this step, includes:

[0017] Continuous wave Doppler radar is used to detect items to be transported on the conveyor line at a preset radar angle;

[0018] The distance from the edge of the item to be transported to the microwave sensor is calculated using continuous wave phase difference ranging or frequency-modulated continuous wave ranging.

[0019] By filtering out environmental noise using Fast Fourier Transform, the characteristic frequencies reflected by the items to be transported are extracted, and real-time dimensions and positioning data of the items to be transported are obtained.

[0020] Optionally, the multi-source detection unit includes an infrared thermal imager and an infrared ranging sensor; the various data of the items to be transported includes the infrared outline of the items to be transported, and the distance data from the left and right edges of the items to be transported to the center line of the conveyor line or a preset position; the step of detecting the items to be transported on the conveyor line based on the multi-source detection unit in response to the start of the conveyor line to obtain various data of the items to be transported, and the steps prior to this step, include:

[0021] On the same side as the high-definition camera, an infrared thermal imager is installed above the transmission line. Multiple infrared thermal imagers are used to form a multi-view mode to provide additional parallax information. The ranging is switched by binocular combination, and infrared data of the items to be transported is obtained by temperature difference recognition.

[0022] Based on the infrared ranging sensor emitting near-infrared rays, the distance to the infrared object to be transported is detected by triangulation or time-of-flight principle.

[0023] Using an uncooled infrared focal plane array, a preset infrared band is detected, and the temperature difference between the item to be transported and the environment is captured to generate a thermal image.

[0024] The region to be transported is segmented based on the region growing algorithm. The infrared outline of the item to be transported is extracted based on the infrared item to be transported data, and the distance data from the left and right edges of the item to be transported to the center line of the transport line or the preset positioning is calculated.

[0025] Optionally, the process of performing multi-source fusion processing on the various items to be transported data according to preset weights to achieve compatible identification of items to be transported of various specifications and materials, in order to obtain the location fusion data of the items to be transported, includes:

[0026] The data of the various items to be transported are uniformly converted into a coordinate system based on the conveyor line;

[0027] Real-time monitoring of ambient temperature, cargo transparency, and light intensity; dynamic weighting of different sensor data based on environmental adaptability and scene adaptation results.

[0028] Based on the dynamic weights, the data of various items to be transported under the coordinate system are fused and calculated to eliminate the error of a single sensor and determine the core parameters of the items to be transported.

[0029] Optionally, the process of determining the positioning direction of the item to be transported based on the fused location data and the preset positioning position includes:

[0030] When comparing the size detection values ​​of the various items to be transported from the data, if the data deviation is greater than the preset size deviation value, the sensor data with the highest weight is selected as the size detection value of the item to be transported.

[0031] Using the centerline of the conveyor line as a reference, the deviation between the center of the item to be conveyed and the centerline of the conveyor line or the preset positioning is calculated by fusing the edge coordinates of the laser, the geometric center of vision, and the edge distance data of microwave or infrared.

[0032] Based on dual-sided multi-point detection using laser sensors, combined with the Hough transform results from vision, the angle between the plane of the item to be transported and the conveyor line is calculated.

[0033] Based on image information captured by a high-definition camera, the system outputs the dimensions of the cargo that extend beyond the edge of the items to be transported and identifies information about fragile items.

[0034] Optionally, the power adjustment unit includes a servo motor, a push rod, a first flexible contact pad, and a position encoder; the process of driving the power adjustment unit to push the item to be transported to the preset positioning position according to the positioning direction of the item to be transported includes:

[0035] The servo motor drives the push rod, which, in conjunction with the first flexible contact pad, pushes the item to be transported to the preset positioning position in real time according to the calculated offset.

[0036] The position encoder is built into the push rod and is used to provide feedback on the push rod's stroke.

[0037] And / or,

[0038] The power adjustment unit further includes a baffle motor, a baffle, and a second flexible contact pad; the process of driving the power adjustment unit to push the item to be transported to the preset positioning position according to the positioning direction of the item to be transported includes:

[0039] The baffle motor controls the lifting and lowering of the baffle, which, together with the second flexible contact pad, ensures that the front end of the item to be conveyed is flush with the front end.

[0040] When the item to be transported comes to a stop, the baffle is mechanically stopped to eliminate the effects of tilting.

[0041] Optionally, the process of driving the power adjustment unit to push the item to be transported to the preset positioning position according to the positioning direction of the item to be transported further includes:

[0042] Based on the integrated core parameters, the distance that the push rod on one side needs to move is calculated for the center of the conveyor line or the preset positioning offset, so as to obtain the basic offset correction displacement.

[0043] If the item to be transported has a tilt angle, the tilt compensation amount is calculated based on the tilt angle and the length of the item to be transported, and the tilt side push rod is pushed forward by the tilt compensation amount to eliminate the tilt through the differential movement of the push rods on both sides.

[0044] If the goods are detected to be beyond the edge of the items to be transported, the movement distance of the side push rod beyond the edge is reduced accordingly, so that the other side push rod can bear the main adjustment.

[0045] Optionally, the process of driving the power adjustment unit to push the item to be transported to the preset positioning position according to the positioning direction of the item to be transported, and the subsequent process, further includes:

[0046] The calculated target displacement and adjustment speed are converted into control signals for the servo motor or baffle motor to drive the push rod or baffle to move.

[0047] The pressure threshold of the push rod or baffle is output synchronously to monitor overload risk in real time;

[0048] The system tracks the location of items to be transported in real time, forming a closed-loop control for adjusting the location of the items to be transported, thus ensuring the final positioning effect.

[0049] This application also provides a dynamic positioning system for conveyor items based on multimodal ranging, including:

[0050] The multi-source detection unit, in response to the start of the conveyor line, detects the items to be conveyed on the conveyor line to obtain various data on the items to be conveyed;

[0051] The data fusion unit performs multi-source fusion processing on the data of the various items to be transported according to preset weights, so as to achieve compatible identification of items to be transported of various specifications and materials, and obtain the location fusion data of the items to be transported.

[0052] The positioning calculation unit determines the positioning direction of the item to be transported based on the fused location data of the item to be transported and the preset positioning location;

[0053] The control and monitoring unit, based on the positioning direction of the item to be transported, drives the power adjustment unit to push the item to be transported to the preset positioning position.

[0054] Optionally, the multi-source detection unit includes a laser sensor; the laser sensors are symmetrically installed on both sides of the conveyor line, with two sets installed symmetrically on each side at a distance of 300mm from left to right.

[0055] Optionally, the multi-source detection unit includes a high-definition camera; the high-definition camera is installed at a height of 2.5m and tilted at 45° above the conveyor line, forming a three-lens high-definition visual inspection system.

[0056] Optionally, the multi-source detection unit includes an infrared thermal imager; the infrared thermal imager and the high-definition camera are mounted on the same side above the conveyor line.

[0057] Optionally, the multi-source detection unit includes a microwave sensor; the microwave sensor is mounted at a beam angle of 30° on both sides of the bottom of the conveyor line and parallel to the conveyor line.

[0058] This application utilizes multimodal ranging to address the issue of glare from metallic items being transported, thereby improving positioning accuracy. Infrared ranging increases the reflectivity of transparent media by 80%, reducing false positives and enabling compatible identification of items of various sizes and materials. It effectively improves positioning success rates even in harsh environments. Passive infrared detection reduces nighttime supplementary lighting energy consumption and avoids interference from supplementary lighting reflections, thus improving detection accuracy. Furthermore, a redundancy mechanism is formed through multi-sensor fusion, automatically switching when a single sensor fails, eliminating downtime risks and significantly enhancing system practicality. Attached Figure Description

[0059] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application. To more clearly illustrate the technical solutions of the embodiments of this application, the drawings used in the description of the embodiments will be briefly introduced below. Obviously, those skilled in the art can obtain other drawings based on these drawings without any creative effort.

[0060] Figure 1 This is a schematic flowchart of a method for dynamic positioning of items on a conveyor line based on multimodal ranging, according to an embodiment of this application.

[0061] Figure 2 This is a schematic diagram of a conveyor line architecture according to an embodiment of this application.

[0062] Figure 3 This is a flowchart illustrating the execution of a multimodal ranging-based dynamic positioning system for conveyor items according to an embodiment of this application.

[0063] Among them, 101, laser sensor group; 102, servo motor; 103, push rod; 104, first flexible contact pad; 105, conveyor line; 107, high-definition camera; 108, image processing module; 109, first pressure sensor; 110, microwave sensor; 111, signal processing module; 112, infrared thermal imager; 113, infrared ranging sensor; 115, baffle; 116, second flexible contact pad; 118, item to be conveyed; 119, forklift; 120, infrared thermal imager processing module.

[0064] The realization of the objectives, functional features, and advantages of this application will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. The accompanying drawings have illustrated specific embodiments of this application, which will be described in more detail below. These drawings and textual descriptions are not intended to limit the scope of the concept in any way, but rather to illustrate the concepts of this application to those skilled in the art through reference to specific embodiments. Detailed Implementation

[0065] It should be understood that the specific embodiments described herein are merely illustrative of this application and are not intended to limit this application.

[0066] Various embodiments of the present application will now be described with reference to the accompanying drawings. In the following description, suffixes such as “module,” “part,” or “unit” used to denote elements are used only for the convenience of the description and have no specific meaning in themselves.

[0067] First Embodiment

[0068] This application proposes a method and system for dynamic positioning of items on a conveyor line based on multimodal ranging. Figure 1 This is a schematic flowchart of a method for dynamic positioning of items on a conveyor line based on multimodal ranging, according to an embodiment of this application.

[0069] like Figure 1 As shown, in one embodiment, the dynamic positioning method for multi-specification items on a conveyor line based on multimodal ranging includes:

[0070] S10: In response to the start of the conveyor line, the items to be conveyed on the conveyor line are detected by the multi-source detection unit to obtain various data on the items to be conveyed.

[0071] For example, the system used in this embodiment can be an automated system for the dynamic positioning and adjustment of items to be transported. The system consists of four parts: a multi-source detection unit, a power adjustment unit, a control and monitoring unit, and the object being monitored / transported. Each unit works together to achieve a closed loop of "detection-judgment-adjustment". Through multi-source sensor fusion detection technology, the system can measure the size of the items to be transported on the conveyor line, determine their tilt, and calculate their positional offset. The power unit adjusts the system in real time to ensure that the items to be transported are accurately positioned or aligned, adapting to the material transfer needs in high-speed and complex environments (such as automated production lines and cold chain warehousing).

[0072] For example, by using four types of sensors—laser, vision, microwave, and infrared—to work together, the detection of features of items to be transported can be covered in different environments, eliminating the limitations of a single sensor.

[0073] S20: Perform multi-source fusion processing on the data of the various items to be transported according to preset weights to achieve compatible identification of items to be transported of various specifications and materials, so as to obtain the location fusion data of the items to be transported.

[0074] For example, the control cabinet (i.e., the PLC) integrates data from multiple sources, performs calculations, determines output parameters, and drives the power unit. Based on environmental adaptability and scenario suitability assessments, the control cabinet assigns dynamic weights to different sensor data, ensuring that core data is prioritized. The control cabinet then fuses the data based on these weights, eliminating errors from individual sensors.

[0075] S30: Determine the positioning direction of the item to be transported based on the fusion data of the item's location and the preset positioning location.

[0076] For example, the control cabinet determines the core parameters of the items to be transported based on the fused data, providing a basis for subsequent displacement and velocity calculations.

[0077] S40: According to the positioning direction of the item to be transported, the drive power adjustment unit pushes the item to be transported to the preset positioning position.

[0078] For example, based on the fused core parameters, the control cabinet calculates the distance that the push rods on both sides need to move, drives the push rods and baffles to position the items to be transported, and ensures that the items to be transported are finally in place and that tilting is eliminated.

[0079] Optionally, the multi-source detection unit includes a laser sensor; the various data on the items to be transported includes the size, edge coordinates, and tilt angle of the items to be transported; the step of detecting the items to be transported on the conveyor line based on the multi-source detection unit in response to the start of the conveyor line, and prior to this step, includes:

[0080] The laser sensor is driven to emit a continuous laser beam, which scans the edge of the item to be transported in real time, causing the edge of the item to block the laser and form diffuse reflection.

[0081] The system receives reflected signals, calculates the distance to the item to be transported using triangulation or time-of-flight methods, and outputs the size, edge coordinates, and tilt angle of the item to be transported.

[0082] For example, two 650nm red laser sensors (accuracy ±0.5mm) are symmetrically installed on each side of the conveyor line (300mm spacing) to detect the length, width, height, and tilt angle of the items to be conveyed. The edges of the items are scanned in real time using a triangulation method. During operation, a continuous laser beam is emitted. The edges of the items block the laser, causing diffuse reflection. The laser sensor receiver receives the reflected signal and calculates the distance, outputting the length, width, and height dimensions of the items (resolution 0.1-1mm), edge coordinates, and tilt angle (detection range ±10°). Alternatively, the distance to the items can be calculated using the Time-of-Flight (ToF) principle, outputting the dimensions of the items (resolution 1-5mm). When using TOF ranging, the laser sensors on each side are asymmetrically installed to avoid receiving additional laser beam interference. For highly reflective items, a built-in polarization filter module reduces reflective interference from metal surfaces (reflectivity attenuation to below 15%).

[0083] Optionally, the multi-source detection unit includes a high-definition camera; the various data on the items to be transported include the geometric center of the items to be transported, the distance from the edge contour of the items to be transported to the centerline of the conveyor line or a preset position; the step of detecting the items to be transported on the conveyor line based on the multi-source detection unit in response to the start of the conveyor line to obtain various data on the items to be transported, and the steps prior to this include:

[0084] A multi-view high-definition visual inspection system is formed by using multiple high-definition cameras at preset shooting heights, distributed at different positions and tilted at preset shooting angles to shoot the top of the items to be transported.

[0085] The edge contour of the item to be transported is extracted by an edge detection algorithm, and the geometric center of the item to be transported and the distance from the edge contour of the item to the centerline of the transport line or the preset position are calculated by combining Hough transform.

[0086] For example, the visual recognition unit includes a high-definition camera (107) and an image processing module (108) to identify the geometric center of the item to be transported and the extent of the goods exceeding the limit.

[0087] Employing two or three 12-megapixel global shutter cameras (30fps) with 8mm fixed-focus lenses, mounted at a height of 2.5m and tilted at 45° to capture images of the top of the item to be transported, a dual / tri-lens high-definition visual inspection system is constructed. Continuous ranging is ensured through "binocular combination switching" (left-center, center-right).

[0088] The image processing module has a built-in GPU acceleration unit, supporting real-time semantic segmentation (distinguishing between the item to be transported and the cargo), and can identify the size of the cargo that exceeds the edge of the item to be transported (minimum detection 5mm). The edge contour of the item to be transported is extracted through the "edge detection algorithm" (Canny operator), and the geometric center of the item to be transported is calculated by combining the "Hough transform" (positioning accuracy ±2mm) and the distance from the edge contour to the centerline of the conveyor line or the preset positioning.

[0089] Optionally, the multi-source detection unit includes a microwave sensor; the various data on the items to be transported includes real-time dimensions and positioning data of the items to be transported; the step of detecting the items to be transported on the conveyor line based on the multi-source detection unit in response to the start of the conveyor line, and prior to this step, includes:

[0090] Continuous wave Doppler radar is used to detect items to be transported on the conveyor line at a preset radar angle;

[0091] The distance from the edge of the item to be transported to the microwave sensor is calculated using continuous wave phase difference ranging or frequency-modulated continuous wave ranging.

[0092] By filtering out environmental noise using Fast Fourier Transform, the characteristic frequencies reflected by the items to be transported are extracted, and real-time dimensions and positioning data of the items to be transported are obtained.

[0093] For example, the microwave detection unit includes a microwave sensor and a signal processing module. The microwave sensor is installed at the bottom of both sides of the conveyor line and emits a 10 GHz signal, which can penetrate dust / water mist and detect the edges of metal items to be conveyed.

[0094] The microwave sensor can employ a 10GHz continuous wave Doppler radar (output power 10mW), with a beam angle of 30°, installed on both sides of the conveyor line parallel to the items to be transported. The distance from the edge of the item to the sensor is calculated using either continuous wave phase difference ranging or frequency modulated continuous wave (FMCW) ranging (output accuracy: 1mm). The microwave signal can penetrate dust (particle size <100μm), water mist (humidity <90%), and non-metallic coverings (such as tarpaulins and cardboard boxes). It has a reflectivity >90% for metallic items to be transported, with a detection size error of ±3mm (unaffected by reflection). However, when metallic goods cover the items to be transported, multipath reflection can increase the error to ±15mm, requiring correction using infrared data.

[0095] The signal processing module uses Fast Fourier Transform (FFT) to filter environmental noise (such as interference from conveyor motors), extracts the characteristic frequencies (1-5kHz) reflected by the items to be transported, and outputs real-time dimensions and positioning data of the items to be transported (updated at a frequency of 100Hz).

[0096] Optionally, the multi-source detection unit includes an infrared thermal imager and an infrared ranging sensor; the various data of the items to be transported includes the infrared outline of the items to be transported, and the distance data from the left and right edges of the items to be transported to the center line of the conveyor line or a preset position; the step of detecting the items to be transported on the conveyor line based on the multi-source detection unit in response to the start of the conveyor line to obtain various data of the items to be transported, and the steps prior to this step, include:

[0097] On the same side as the high-definition camera, an infrared thermal imager is installed above the transmission line. Multiple infrared thermal imagers are used to form a multi-view mode to provide additional parallax information. The ranging is switched by binocular combination, and infrared data of the items to be transported is obtained by temperature difference recognition.

[0098] Based on the infrared ranging sensor emitting near-infrared rays, the distance to the infrared object to be transported is detected by triangulation or time-of-flight principle.

[0099] Using an uncooled infrared focal plane array, a preset infrared band is detected, and the temperature difference between the item to be transported and the environment is captured to generate a thermal image.

[0100] The region to be transported is segmented based on the region growing algorithm. The infrared outline of the item to be transported is extracted based on the infrared item to be transported data, and the distance data from the left and right edges of the item to be transported to the center line of the transport line or the preset positioning is calculated.

[0101] For example, in the infrared detection unit, the infrared thermal imager can be installed above the conveyor line (on the same side as the camera) to obtain infrared data of the item to be conveyed (there is a temperature difference of 2-5℃ between the item to be conveyed and the environment) by temperature difference recognition.

[0102] The infrared ranging sensor emits 940nm near-infrared light (modulation frequency 100kHz) and uses the principles of "triangulation" or "time-of-flight (ToF)" to detect distance (range 0.5-5m, accuracy ±1mm). Infrared light has a reflectivity >80% in transparent media (such as plastic film and glass), allowing it to penetrate transparent films 0.1mm-2mm thick, directly detecting the edge of the item to be transported, thus eliminating misjudgments caused by laser / microwave penetration of transparent goods (error reduced to ±2mm).

[0103] Infrared thermal imager processing module (120): Employs two or three thermal imagers in a dual / trino mode. The third imager provides additional parallax information. When the dual-view system fails due to large-area obstruction (e.g., severe damage to the edge of the item to be transported), continuous ranging can be ensured through "dual-view combination switching" (left-center, center-right). An uncooled infrared focal plane array (640×512 resolution) is used, with a detection band of 8-14μm. Thermal images are generated by capturing the temperature difference (≥2℃) between the item to be transported and the environment. The area of ​​the item to be transported is segmented based on a "region growing algorithm," and the infrared edge contour is extracted (positioning accuracy ±3mm). The distance from the left and right edges of the item to be transported to the centerline is calculated. In a cold chain environment of -10℃ to 5℃, it can penetrate 5mm thick fog (dew point temperature below 0℃), solving the low-temperature attenuation problem of laser / microwave.

[0104] Optionally, the process of performing multi-source fusion processing on the various items to be transported data according to preset weights to achieve compatible identification of items to be transported of various specifications and materials, in order to obtain the location fusion data of the items to be transported, includes:

[0105] The data of the various items to be transported are uniformly converted into a coordinate system based on the conveyor line;

[0106] Real-time monitoring of ambient temperature, cargo transparency, and light intensity; dynamic weighting of different sensor data based on environmental adaptability and scene adaptation results.

[0107] Based on the dynamic weights, the data of various items to be transported under the coordinate system are fused and calculated to eliminate the error of a single sensor and determine the core parameters of the items to be transported.

[0108] For example, the control cabinet first receives the raw data from each sensor and performs preliminary cleaning and format standardization to ensure data validity. During the preprocessing, obvious outliers (such as out-of-range data caused by sensor malfunction) are removed, and all data is uniformly converted into a coordinate system based on the centerline of the conveyor line (to facilitate subsequent offset calculation).

[0109] Based on the results of environmental adaptability and scene adaptation assessments, the control cabinet dynamically assigns weights to different sensor data to ensure that core data is prioritized. In the environmental adaptability assessment, it monitors the stability of the laser signal (deviation > 10mm for three consecutive tests) and the clarity of the visual image in real time; in harsh environments (such as dust or glare), the microwave detection unit is activated as the primary detection source. During the scene adaptation assessment, the control unit monitors the ambient temperature (≤5℃ triggers cold chain mode), cargo transparency (laser penetration > 70% triggers transparent mode), and light intensity (<200 lux triggers low-light mode); in the corresponding mode, it automatically activates the infrared detection unit and adjusts the weights of the multi-source data.

[0110] For example, in a normal temperature, dry environment with normal lighting, laser (60% weight) and vision (30% weight) are the main methods, with microwave and infrared as redundancy (10% weight), taking advantage of the high precision of laser and the geometric center confirmation of vision.

[0111] For example, in environments with metal items to be transported, high reflectivity, dust, or water mist, microwave is activated as the main detection source (70% weight), with laser (polarization filtering enabled, 20% weight) and infrared (10% weight) as auxiliary sources to counteract reflection and occlusion interference.

[0112] For example, in a cold chain environment, such as around -5 degrees Celsius, there will be a lot of fog. An infrared thermal imager (weight 60%) can be used as the core, with microwaves (weight 30%) as an auxiliary, to correct the low-temperature attenuation error of lasers / microwaves.

[0113] For example, in the case of transparent cargo coverage, an infrared ranging sensor (weight 80%) can be used as the primary sensor, with visual contours (weight 20%) as the auxiliary sensor, to avoid misjudgments caused by laser / microwave penetration of the transparent medium.

[0114] For example, in low-light environments, an infrared thermal imager (70% weighting) is used instead of vision, supplemented by microwaves (30% weighting) to avoid interference from supplemental lighting and reflections.

[0115] For example, the control cabinet fuses data based on weights to eliminate errors from individual sensors, and finally fuses the data to determine the core parameters of the item to be transported, providing a basis for subsequent calculations of the displacement and speed of push rods and baffles.

[0116] Optionally, the process of determining the positioning direction of the item to be transported based on the fused location data and the preset positioning position includes:

[0117] When comparing the size detection values ​​of the various items to be transported from the data, if the data deviation is greater than the preset size deviation value, the sensor data with the highest weight is selected as the size detection value of the item to be transported.

[0118] Using the centerline of the conveyor line as a reference, the deviation between the center of the item to be conveyed and the centerline of the conveyor line or the preset positioning is calculated by fusing the edge coordinates of the laser, the geometric center of vision, and the edge distance data of microwave or infrared.

[0119] Based on dual-sided multi-point detection using laser sensors, combined with the Hough transform results from vision, the angle between the plane of the item to be transported and the conveyor line is calculated.

[0120] Based on image information captured by a high-definition camera, the system outputs the dimensions of the cargo that extend beyond the edge of the items to be transported and identifies information about fragile items.

[0121] For example, when comparing the detected dimensions of the item to be transported using laser, microwave, and infrared sensors, if the deviation of a certain sensor's data from other data is greater than 30mm, the sensor data with the highest weight is used (e.g., microwave data is used for metal items to be transported). For instance, if the laser detection width is 1215mm (error + 15mm), the microwave detection width is 1202mm (error + 2mm), and the infrared auxiliary verification shows no abnormalities, then the fused width is 1202mm.

[0122] For example, when determining the center offset of the item to be transported, the centerline of the transport line is used as a reference. The deviation (in mm) between the center of the item and the centerline of the transport line or a preset position is calculated by fusing the edge coordinates of the laser, the geometric center of the vision, and the edge distances of the microwave / infrared sensors. During the calculation, half the difference between the distance from the left edge of the item to the preset position and the distance from the right edge of the item to the preset position is taken as the center offset. For example, if the left edge is 611 mm from the preset position after fusion, and the right edge is 589 mm from the preset position, then the preset position offset ΔX = (611-589) / 2 = +11 mm (i.e., it needs to be offset 11 mm to the right).

[0123] For example, based on dual-sided multi-point detection (2 sets on each side, 300mm spacing) using a laser sensor, combined with the Hough transform results from vision, the angle (range ±10°) between the plane of the item to be transported and the transport line is calculated as the tilt angle of the item to be transported. For instance, if the laser detection shows a left front offset of -10mm and a rear offset of 0mm, and a right front offset of +5mm and a rear offset of +15mm, the fused tilt angle θ = 5.2°.

[0124] For example, regarding the data on the status of the goods, the visual recognition unit outputs information such as the size of the goods exceeding the edge of the item to be transported (e.g., exceeding 50mm on the left) and whether there are fragile items, as constraints for subsequent speed and thrust control.

[0125] Optionally, the power adjustment unit includes a servo motor, a push rod, a first flexible contact pad, and a position encoder; the process of driving the power adjustment unit to push the item to be transported to the preset positioning position according to the positioning direction of the item to be transported includes:

[0126] The servo motor drives the push rod, which, in conjunction with the first flexible contact pad, pushes the item to be transported to the preset positioning position in real time according to the calculated offset.

[0127] The position encoder is built into the push rod and is used to provide feedback on the push rod's stroke.

[0128] Figure 2 This is a schematic diagram of a conveyor line according to an embodiment of this application.

[0129] like Figure 2 As shown, exemplarily, in the power adjustment component of the power adjustment unit, the servo motor 102 drives the push rod 103, which, in conjunction with the first flexible contact pad 104 (to avoid damaging the item to be transported), pushes the item to be transported to the preset positioning position in real time according to the preset positioning offset; the built-in position encoder provides feedback on the push rod stroke to ensure adjustment accuracy of ±1mm.

[0130] Optionally, the power adjustment unit further includes a baffle motor, a baffle, and a second flexible contact pad; the process of driving the power adjustment unit to push the item to be transported to the preset positioning position according to the positioning direction of the item to be transported includes:

[0131] The baffle motor controls the lifting and lowering of the baffle, which, together with the second flexible contact pad, ensures that the front end of the item to be conveyed is flush with the front end.

[0132] When the item to be transported comes to a stop, the baffle is mechanically stopped to eliminate the effects of tilting.

[0133] Please continue to refer to this. Figure 2 For example, in the flush positioning component of the power adjustment unit, the baffle motor 114 controls the baffle 115 to rise and fall, and works with the second flexible contact pad 116 to achieve flush positioning of the front end of the item to be conveyed (suitable for single-point detection scenarios); when the item to be conveyed stops, the baffle mechanically limits the movement to eliminate the tilting effect.

[0134] For example, the system is also equipped with pressure sensors to detect the thrust of the push rod / baffle. When the force exceeds the threshold (e.g., 300N / 500N), an alarm is triggered to avoid overload damage to the equipment or the items to be transported and to identify abnormal situations in a timely manner.

[0135] Optionally, the process of driving the power adjustment unit to push the item to be transported to the preset positioning position according to the positioning direction of the item to be transported further includes:

[0136] Based on the fused core parameters, the distance that the push rod on one side needs to move is calculated for the preset positioning offset, so as to obtain the basic offset correction displacement.

[0137] If the item to be transported has a tilt angle, the tilt compensation amount is calculated based on the tilt angle and the length of the item to be transported, and the tilt side push rod is pushed forward by the tilt compensation amount to eliminate the tilt through the differential movement of the push rods on both sides.

[0138] If the goods are detected to be beyond the edge of the items to be transported, the movement distance of the side push rod beyond the edge is reduced accordingly, so that the other side push rod can bear the main adjustment.

[0139] For example, based on the integrated core parameters, the control cabinet calculates the distance that the push rods on both sides need to move to ensure that the items to be transported are finally in place and that tilting is eliminated.

[0140] During the basic offset correction displacement process, for the preset positioning offset ΔX, the distance that the single-sided push rod needs to move is |ΔX| (e.g., if ΔX=+15mm, the right push rod needs to be pushed 15mm to the left).

[0141] In tilt correction, if the item to be transported has a tilt angle θ, the tilt needs to be eliminated by the differential movement of the push rods on both sides: the push rod on the tilted side (such as the left side forward) advances an additional "tilt compensation amount" (calculated based on the tilt angle and the length of the item to be transported, usually 5-20mm). For example, for a narrow item to be transported (800mm×600mm) tilted at 5°, the right push rod advances 25mm to correct the offset, and the left push rod advances 10mm simultaneously to correct the tilt.

[0142] During cargo clearance and adjustment, if cargo extends beyond the edge of the item to be transported, the movement distance of the side push rod should be reduced (to avoid crushing the cargo). For example, if the cargo on the left side extends 50mm beyond the edge, only advance it 5mm (normally it needs to be advanced 22mm), and the right push rod will handle the main adjustment.

[0143] For example, the conveyor belt speed setting needs to balance efficiency and safety. The control cabinet can dynamically select the speed based on the status of the items to be conveyed, the environmental scenario, and the characteristics of the goods. For instance, in a normal temperature environment with standard empty items to be conveyed, the speed can be efficiently adjusted to 50 mm / s. When the conveyor belt is tilted or the goods are overloaded, the speed can be reduced to 20-30 mm / s to avoid goods shaking or jamming. For heavy metal items to be conveyed, the speed can be adjusted to 30-40 mm / s to accommodate thrust requirements. For identified cold chain and fragile items to be conveyed, the speed should not exceed 20 mm / s, employing a low-impact adjustment strategy. In harsh environments such as dusty or water mist environments, the speed can be reduced by 10-20%. Real-time monitoring and correction can be used to avoid over-adjustment.

[0144] Optionally, the process of driving the power adjustment unit to push the item to be transported to the preset positioning position according to the positioning direction of the item to be transported, and the subsequent process, further includes:

[0145] The calculated target displacement and adjustment speed are converted into control signals for the servo motor or baffle motor to drive the push rod or baffle to move.

[0146] The pressure threshold of the push rod or baffle is output synchronously to monitor overload risk in real time;

[0147] The system tracks the location of items to be transported in real time, forming a closed-loop control for adjusting the location of the items to be transported, thus ensuring the final positioning effect.

[0148] For example, the control cabinet converts the calculated "target displacement" and "adjustment speed" into control signals for the servo motor, such as pulse count or analog signals, to drive the push rod. Simultaneously, it outputs the threshold value of the pressure sensor (e.g., 300N) for real-time monitoring of overload risk. Throughout the process, the control cabinet tracks the position of the item to be transported in real time using a position encoder and multi-source sensors. If the deviation is not within the acceptable range (e.g., greater than 3mm), the data is returned for recalculation, forming a closed-loop control of "detection-calculation-adjustment-feedback" to ensure that the final positioning accuracy is no greater than ±3mm.

[0149] Second Embodiment

[0150] Based on Example 1, please continue to refer to... Figure 2 This application also provides a dynamic positioning system for conveyor items based on multimodal ranging, comprising:

[0151] The multi-source detection unit, in response to the start of the conveyor line, detects the items to be conveyed on the conveyor line to obtain various data on the items to be conveyed;

[0152] The data fusion unit performs multi-source fusion processing on the data of the various items to be transported according to preset weights, so as to achieve compatible identification of items to be transported of various specifications and materials, and obtain the location fusion data of the items to be transported.

[0153] The positioning calculation unit determines the positioning direction of the item to be transported based on the fused location data of the item to be transported and the preset positioning location;

[0154] The control and monitoring unit, based on the positioning direction of the item to be transported, drives the power adjustment unit to push the item to be transported to the preset positioning position.

[0155] Please continue to refer to this. Figure 2 The following explanation will take pallet 118 conveyor as an example.

[0156] Optionally, the multi-source detection unit includes a laser sensor 101; the laser sensor 101 is symmetrically installed on both sides of the conveyor line 105, with two sets installed symmetrically on each side at a left-right spacing of 300mm.

[0157] In one embodiment, for the multi-source detection unit, laser sensors 101 can have laser emitters symmetrically distributed on both sides of the conveyor line 105, two on each side (300mm apart), with the lenses pointing towards the edge of the tray 118. These sensors are equipped with a 650nm red laser, with an accuracy of ±0.5mm, a detection range of ±10° tilt angle, and a built-in polarization filter module (anti-reflective). They are used to scan the dimensions and edge coordinates of the tray 118 in real time and output initial positioning data.

[0158] Laser sensor receivers 101_1 are symmetrically distributed on both sides of the conveyor line 105, with two on each side, and the lenses pointing towards the edge of the tray 118. The receiving lens is an 8mm convex lens with a field of view of 30° horizontally and 20° vertically. The photosensitive element is a linear CCD with a minimum detection light power of 10nW. The baseline length (L=50mm) and emission angle (α=30°) of the transmitter and receiver are fixed to ensure that the reflected light always falls on the CCD sensor of the receiver within the normal distance (20~500mm) of the tray 118.

[0159] Optionally, the multi-source detection unit includes a high-definition camera 107; the high-definition camera 107 is installed at a height of 2.5m and tilted at 45° above the conveyor line 105, forming a three-lens high-definition visual inspection system.

[0160] For example, a high-definition camera 107 is positioned 2.5 meters above the conveyor line 105, centered, with the lens tilted at 45° towards the top of the pallet 118. It employs a 12-megapixel camera, global shutter, 30fps frame rate, and an 8mm fixed-focus lens. It is used for visual identification of the geometric center of the pallet 118, detecting goods exceeding the edge, and correcting laser detection errors.

[0161] Optionally, the multi-source detection unit includes a microwave sensor 110; the microwave sensor 110 is mounted at a beam angle of 30° on both sides of the bottom of the conveyor line 105 and parallel to the conveyor line 105.

[0162] For example, microwave sensors 110 are disposed at the bottom of both sides of the conveyor line 105 (100mm from the ground), with the transmitting end pointing towards the bottom of the tray 118. They are equipped with a 10GHz continuous wave Doppler radar with a beam angle of 30°, capable of penetrating dust / water mist (particle size <100μm). This is used for anti-reflective detection of the metal tray 118, replacing laser as the primary detection source in harsh environments.

[0163] Optionally, the multi-source detection unit includes an infrared thermal imager 112; the infrared thermal imager 112 and the high-definition camera 107 are mounted on the same side above the conveyor line 105.

[0164] The infrared thermal imager 112 can be installed above the conveyor line 105, parallel to the camera. It is configured with a 640×512 resolution, 8-14μm wavelength, and a detection temperature difference ≥2℃. It can identify the outline of the tray 118 through temperature difference in cold chain foggy environments, replacing visual inspection in low-light scenes.

[0165] The infrared ranging sensor 113 is mounted parallel to the laser sensor and positioned outside it. It features 940nm near-infrared light, operates on the ToF principle, has an accuracy of ±1mm, and a reflectivity of >80% for transparent media. It can penetrate transparent films / glass to detect the edge of the tray 118, thus resolving misjudgments caused by laser / microwave penetration.

[0166] Infrared ranging sensor receivers 113_1 can be symmetrically distributed on both sides of the conveyor line 105, with two receivers on each side, and the lens pointing towards the edge of the tray 118. The receiving spectral range is 900-980nm, the receiving lens focal length is 6mm, the photosensitive element is a high-speed CMOS array, and the minimum detection light power is 50nW. It can receive infrared reflected beams and measure distance using triangulation. The baseline length (L=40mm) and emission angle (α=35°) of the transmitter and receiver are fixed to ensure that within the normal distance (20~500mm) of the tray 118, the reflected light always falls on the receiver's CMOS sensor.

[0167] The servo motor 102 can be configured on both sides of the conveyor line 105, located outside the laser sensor, and rigidly connected to the push rod. It is configured with a rated torque of 5 N·m, a position control accuracy of ±0.01 mm, and a built-in encoder. It can drive the push rod to complete precise displacement according to PLC commands (speed / force adjustable).

[0168] The push rod 103 is a horizontally elongated strip, with one end connected to a servo motor and the other end extending to the edge of the tray 118. A pressure sensor 109 is integrated in the middle to detect the thrust, with a range of 0-500N. The push rod 103 receives power from the motor and pushes the tray 118 to achieve the desired adjustment.

[0169] The first flexible contact pad is located at the end of the push rod and makes direct contact with pallet 118. It is made of polyurethane elastic material with a hardness of 50 Shore A and a thickness of 10mm. It is used to cushion the pushing force and prevent damage to pallet 118 (especially suitable for aerospace aluminum pallets).

[0170] The baffle motor 114 is located at the lower end of the conveyor line 105 (the front end of the pallet 118) and is rigidly connected to the push rod. It has a rated thrust of 100N and a stroke of 300mm. It drives the push rod to complete the up-and-down displacement of the baffle according to PLC commands (speed / force adjustable).

[0171] The baffle 115 is a horizontal, elongated strip connected to a baffle motor. A pressure sensor 109 (detecting thrust, range 0-500N) is integrated in its center. It receives power from the motor, blocks the tray 118, and allows the head of the tray 118 to be aligned and adjusted.

[0172] The second flexible contact pad 116 is located at the end of the baffle, connected to the baffle, and in direct contact with the pallet 118. It is made of polyurethane elastic material with a hardness of 50 Shore A and a thickness of 10 mm. It is used to cushion the thrust and prevent damage to the pallet 118 (especially suitable for aerospace aluminum pallets).

[0173] The conveyor line 105 uses a horizontal conveyor belt or rollers with a non-slip textured surface (friction coefficient ≥0.8) and a width adaptable to various pallet sizes. It is used to carry and transport pallets 118, providing a stable platform for positioning.

[0174] Control cabinet 106 is an independent unit on the right side of the system, connected to all sensors and motors via wiring. Its core functions are to fuse laser / vision / microwave / infrared data, calculate the displacement and velocity of the push rod target, drive the power unit to operate, and monitor anomalies in real time.

[0175] The system also includes an image processing module 108, integrated with the camera and featuring a built-in GPU acceleration unit. It runs Canny edge detection and Hough transform algorithms to extract the contours and perform semantic segmentation of the tray 118.

[0176] The first pressure sensor 109 is located in the middle of the push rod and connected in series in the power transmission path. It is used to detect the push rod thrust and identify jamming or overload (an alarm is triggered when the threshold is >300N).

[0177] The second pressure sensor 117 is integrated into the baffle push rod, located in the middle of the push rod, and connected in series in the path between the baffle and the push rod. It is used to detect the baffle pressure and identify jamming or overload (an alarm is triggered when the threshold is >500N).

[0178] The infrared thermal imager processing module 120 is located below the camera. It has a thermal image frame rate of 30Hz (real-time output), can identify low temperature differences, and extract thermal contours in real time. It is used to process infrared thermal images and calculate the distance from the edge of the tray 118 to the centerline of the conveyor line 105 or a preset position.

[0179] Figure 3This is a flowchart illustrating the execution of a multimodal ranging-based dynamic positioning system for conveyor items according to an embodiment of this application.

[0180] like Figure 3 As shown, in one embodiment, the conveyor line item dynamic positioning system based on multimodal ranging performs the following steps during operation:

[0181] Step 1: Laser sensor detection (obtain initial width, offset, and tilt angle).

[0182] Step 1.2: Microwave-assisted detection

[0183] The microwave sensor (110) penetrates the interference and detects the actual width of the object (compared with the laser data; if the deviation is >30mm, the microwave data shall prevail).

[0184] In scenarios involving metal objects, correct laser errors caused by reflections.

[0185] Step 1.3: Infrared-assisted correction

[0186] Cold chain mode: The infrared thermal imager (112) generates a temperature profile map, extracts the edge of the product, and corrects the width detection error caused by fog in the laser (typical correction value 10-30mm).

[0187] Transparent mode: The infrared ranging sensor (113) emits 940nm infrared light to detect the actual width of the object under the transparent film, avoiding false judgments due to laser penetration;

[0188] Low-light mode: The infrared thermal imager replaces visual recognition, outputting the outline of the object without supplemental lighting, and calculates the preset positioning offset by fusing it with microwave data.

[0189] Step 1.5: Visual recognition correction (confirming the geometric center and cargo exceeding the permitted limits).

[0190] Step 2: Control cabinet fusion calculation (based on multi-source data, the displacement and velocity of the push rod target are calculated by fusing data from various items).

[0191] Step 3: Adjust the push rod (the servo motor drives the push rod movement).

[0192] Step 3.5: Real-time monitoring (laser + vision + microwave + infrared collaborative tracking, dynamic correction of actions).

[0193] Step 4: Compliance judgment (if the deviation is ≤3mm, the process is complete; otherwise, return to step 2).

[0194] Step 5: Anomaly detection (if the thrust exceeds the threshold or the displacement is abnormal, an alarm will be triggered for protection).

[0195] The following examples illustrate the dynamic positioning system for conveyor items based on multimodal ranging in several common implementation scenarios.

[0196] Example 1 of implementation scenario: standard empty pallet moves to preset location

[0197] When a standard 1200mm × 1000mm pallet enters the conveyor line:

[0198] Scenario: Normal temperature and dry, light intensity 300-5000 lux, dust <5mg / ㎡.

[0199] Laser sensor 101 detects a width of 1200mm, with a preset positioning offset of +30mm (offset to the right), and no tilt.

[0200] The visual recognition unit confirmed a preset positioning offset of +30mm, and no goods exceeded the offset.

[0201] Control cabinet 106 calculation: The left push rod remains stationary, while the right push rod advances 30mm at a speed of 50mm / s;

[0202] After the push rod action is completed, the position encoder reports that the displacement has met the target, and the preset positioning movement is complete.

[0203] Example Scenario 2: Narrow pallet with tilt

[0204] When the 800mm×600mm pallet is tilted at 5° (left side forward):

[0205] Scenario: Normal temperature and dry, light intensity 300-5000 lux, dust <5mg / ㎡.

[0206] The laser sensor detects a width of 800mm, with a preset positioning offset of -25mm (offset to the left) and a tilt of 5°.

[0207] The visual recognition correction angle is 5.2°, and no goods exceed the limit;

[0208] Adjustment strategy: Advance the right push rod by 25mm (correct offset), and simultaneously advance the left push rod by 10mm (correct tilt), reducing the speed to 30mm / s;

[0209] The camera monitors the process in real time, and stops once the tray posture is confirmed to be corrected.

[0210] Example 3 of implementation scenario: Goods exceeding the limit

[0211] When the goods on the left side of a 1100mm x 1100mm pallet extend 50mm beyond the edge:

[0212] Scenario: Normal temperature and dry, light intensity 300-5000 lux, dust <5mg / ㎡.

[0213] The laser sensor experienced data interruption on the left side due to obstruction by cargo, and only detected the right edge.

[0214] The visual recognition unit identifies the actual width of the item as 1100mm, with a preset positioning offset of +22mm, and the item on the left side exceeds 50mm.

[0215] Adjustment strategy: Advance the right push rod 22mm, and advance the left push rod only 5mm (to avoid exceeding the cargo), at a speed of 20mm / s;

[0216] Pressure sensor 109 monitors thrust (≤300N) in real time to ensure no overload.

[0217] Implementation Scenario Example 4: Exception Handling (Stuck)

[0218] When an item is stuck due to a foreign object at the bottom and cannot be moved:

[0219] Alarm function prompt: Pressure abnormality warning.

[0220] When the push rod is pushed forward, the pressure sensor 109 detects that the thrust suddenly increases to 600N (exceeding the threshold of 500N).

[0221] Control cabinet 106 immediately stops the push rod action, triggers an audible and visual alarm, and displays "Item stuck" on the screen;

[0222] The system records the location of the abnormality and will restart after the foreign object is manually removed.

[0223] Implementation Scenario Example 5: Sensor Fusion Correction

[0224] When the laser sensor misjudges the width of the object due to reflection (detected value 1300mm, actual 1200mm).

[0225] Alarm function prompt: Laser sensor cleaning reminder.

[0226] The visual recognition unit confirmed the actual width was 1200mm through image analysis and determined that the laser data was abnormal.

[0227] The system switches to vision-driven mode and calculates and adjusts parameters based on a width of 1200mm;

[0228] Simultaneously triggers a laser sensor cleaning prompt (via the PLC control cabinet display).

[0229] Implementation Scenario Example 6: Moving a metal pallet into place in a dusty environment

[0230] Scenario: Aluminum items (1200mm×1000mm) are handled in a freight station with a dust concentration of 8mg / m³;

[0231] After the polarization filter is turned on, the detection width of laser sensor 101 is 1215mm (error + 15mm).

[0232] Environmental assessment activates the microwave detection unit. The microwave sensor 110 penetrates the dust, and with infrared auxiliary monitoring, the actual width is detected as 1202mm (error + 2mm), with a preset positioning offset of -15mm.

[0233] The PLC drives the right push rod to advance 15mm, and the microwave monitors the displacement in real time to complete the positioning (final deviation 1mm).

[0234] Implementation Scenario Example 7: Pallet Positioning Covered by Transparent Plastic Film

[0235] Scene: 1100mm×1100mm wooden pallet covered with 0.5mm transparent PET film;

[0236] The laser sensor penetrated the film layer and misjudged the width by 1300mm (error + 200mm); the microwave sensor simultaneously detected a width of 1295mm (error + 195mm).

[0237] Scene adaptation triggers transparent mode; infrared ranging sensor 113 emits 940nm infrared light; visual contour-assisted monitoring detects the actual width under the film as 1100mm, with a preset positioning offset of +22mm.

[0238] The control unit drives the left push rod to advance 22mm, and the infrared system tracks the film's sliding status in real time, ensuring no collision with goods during the adjustment process.

[0239] Implementation Scenario Example 8: Positioning in a Cold Chain Fog Environment

[0240] Scenario: -5℃ cold chain freight station (5m visibility due to fog), 1000mm×800mm wooden items (temperature -3℃, ambient temperature -5℃, temperature difference 2℃);

[0241] Due to fog scattering, the detection width of the laser sensor is 980mm (error -20mm); due to low-temperature attenuation, the detection width of the microwave sensor is 1015mm (error +15mm).

[0242] The infrared thermal imager 112 identifies contours based on temperature differences, with a detection width of 1001mm (error + 1mm) and a preset positioning offset of + 18mm.

[0243] The PLC uses infrared data as a reference to drive the left push rod forward 18mm. During the process, the infrared data is corrected in real time, and the deviation is ≤2mm after completion.

[0244] Example Scenario 9: Comparative Experiment on Anti-Reflection of Laser and Microwave

[0245] Scene: An aluminum object (surface roughness Ra1.6μm) under strong light (10000 lux).

[0246] When the polarization filter of laser sensor 101 is not enabled, the detection width is 1300mm (actual 1200mm, error +100mm); after enabling the filter, the error drops to +15mm.

[0247] The microwave sensor 110 directly detects a width of 1202mm (error + 2mm), is unaffected by reflection, and the system ultimately completes the movement to the correct position based on the microwave data.

[0248] Compared with the prior art, the beneficial effects of this application are as follows:

[0249] Microwave detection solves the problem of glare on metal objects, with a positioning error of ≤±3mm; infrared ranging improves the reflectivity of transparent media by 80%, reduces the false judgment rate to below 0.5%, and achieves multi-material compatibility.

[0250] Laser, microwave, and infrared technologies work together to penetrate dust, water mist, and low-temperature fog, increasing the positioning success rate to 99% in harsh environments. Passive infrared detection reduces nighttime supplementary lighting energy consumption by 60%, avoids interference from supplementary lighting reflections, and improves adaptability to harsh environments.

[0251] The four-sensor fusion forms a redundancy mechanism, automatically switching when a single sensor fails, eliminating the risk of downtime and achieving 99.9% system availability, thus providing multi-source redundancy protection.

[0252] Dynamic adjustment strategies for goods with cargo (such as flexible contact pads + thrust monitoring) reduce the collision rate to below 0.1%; it is compatible with multiple specifications and materials of goods that can enter the conveyor line, without the need for manual parameter adjustment, thus achieving cargo safety and compatibility with specifications and materials.

[0253] This application utilizes multimodal ranging to address the issue of glare from metallic objects using microwave detection, thereby improving positioning accuracy. Infrared ranging increases the reflectivity of transparent media by 80%, reducing false positives and enabling compatible identification of items of various sizes and materials. It effectively improves positioning success rates even in harsh environments. Passive infrared detection reduces nighttime supplementary lighting energy consumption and avoids interference from supplementary lighting reflections, thus improving detection accuracy. Furthermore, a redundancy mechanism is formed through multi-sensor fusion, automatically switching when a single sensor fails, eliminating downtime risk and significantly enhancing system practicality.

[0254] It should be noted that step designations such as S10 and S20 are used in this application for the purpose of more clearly and concisely describing the corresponding content, and do not constitute a substantial limitation on the order. In specific implementation, those skilled in the art may execute S20 first and then S10, etc., but these should all be within the protection scope of this application.

[0255] In the embodiments of the apparatus and storage medium provided in this application, all the technical features of any of the above-described method embodiments may be included. The extended and explanatory content of the specification is basically the same as that of the embodiments of the above methods, and will not be repeated here.

[0256] This application also provides a computer program product, which includes computer program code. When the computer program code is run on a computer, it causes the computer to perform the methods described in the various possible implementations above.

[0257] This application also provides a chip, including a memory and a processor. The memory is used to store a computer program, and the processor is used to call and run the computer program from the memory, so that a device with the chip installed performs the methods described in the various possible implementations above.

[0258] It is understood that the above scenarios are merely examples and do not constitute a limitation on the application scenarios of the technical solutions provided in the embodiments of this application. The technical solutions of this application can also be applied to other scenarios. For example, as those skilled in the art will know, with the evolution of device architecture and the emergence of new business scenarios, the technical solutions provided in the embodiments of this application are also applicable to similar technical problems.

[0259] The sequence numbers of the embodiments in this application are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0260] The steps in the method of this application embodiment can be adjusted, combined, or deleted according to actual needs.

[0261] The units in the device of this application embodiment can be merged, divided, and deleted according to actual needs.

[0262] In this application, the same or similar terms, concepts, technical solutions and / or application scenario descriptions are generally described in detail only when they appear for the first time. When they appear again, they are generally not repeated for the sake of brevity. When understanding the technical solutions and other contents of this application, the same or similar terms, concepts, technical solutions and / or application scenario descriptions that are not described in detail later can be referred to their previous relevant detailed descriptions.

[0263] In this application, the descriptions of the various embodiments have different focuses. For parts that are not described in detail or recorded in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0264] The technical features of the present application can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of the present application.

[0265] The above are merely preferred embodiments of this application and do not limit the scope of this application. Any equivalent structural or procedural transformations made based on the description and drawings of this application, or direct or indirect applications in other related technical fields, are similarly included within the scope of protection of this application.

Claims

1. A method for dynamic positioning of items on a conveyor line based on multimodal ranging, characterized in that, include: In response to the start of the conveyor line, the multi-source detection unit detects the items to be conveyed on the conveyor line to obtain various data on the items to be conveyed; The data of the various items to be transported are fused from multiple sources according to preset weights to achieve compatible identification of items of various specifications and materials, so as to obtain the location fusion data of the items to be transported. Based on the fused location data of the item to be transported and the preset positioning location, the positioning direction of the item to be transported is determined; According to the positioning direction of the item to be transported, the drive power adjustment unit pushes the item to be transported to the preset positioning position; The multi-source detection unit includes a laser sensor, a high-definition camera, a microwave sensor, an infrared thermal imager, and an infrared ranging sensor. The laser sensor is used to acquire the size, edge coordinates, and tilt angle of the item to be transported. The high-definition camera is used to acquire the geometric center of the item to be transported and the distance from the edge contour of the item to the centerline of the transport line or a preset position. The microwave sensor is used to acquire real-time size and positioning data of the item to be transported. The infrared thermal imager and infrared ranging sensor are used to acquire infrared contour data of the item to be transported and the distance data from the left and right edges of the item to the centerline of the transport line or a preset position. The process of performing multi-source fusion processing on the various types of items to be transported according to preset weights to achieve compatible identification of items of various specifications and materials, and to obtain the location fusion data of the items to be transported, includes: The data of the various items to be transported are uniformly converted into a coordinate system based on the conveyor line; Real-time monitoring of ambient temperature, cargo transparency, and light intensity; dynamic weighting of different sensor data based on environmental adaptability and scene adaptation results. Based on the dynamic weights, the data of various items to be transported under the coordinate system are fused and calculated to eliminate the error of a single sensor and determine the core parameters of the items to be transported.

2. The method for dynamic positioning of items on a conveyor line based on multimodal ranging according to claim 1, characterized in that, The laser sensor is used to acquire the dimensions, edge coordinates, and tilt angle of the item to be transported, including: The laser sensor is driven to emit a continuous laser beam, which scans the edge of the item to be transported in real time, causing the edge of the item to block the laser and form diffuse reflection. Receive the reflected signal, calculate the distance to the item to be transported using triangulation or time-of-flight method, and output the size, edge coordinates, and tilt angle of the item to be transported. And / or, The high-definition camera is used to acquire the geometric center of the item to be transported, the distance from the edge contour of the item to the centerline of the conveyor line, or a preset position, including: A multi-view high-definition visual inspection system is formed by using multiple high-definition cameras at preset shooting heights, distributed at different positions and tilted at preset shooting angles to shoot the top of the items to be transported. The edge contour of the item to be transported is extracted by an edge detection algorithm, and the geometric center of the item to be transported and the distance from the edge contour of the item to the centerline of the transport line or the preset position are calculated by combining Hough transform. And / or, The microwave sensor is used to acquire real-time dimensions and positioning data of the item to be transported, including: Continuous wave Doppler radar is used to detect items to be transported on the conveyor line at a preset radar angle; The distance from the edge of the item to be transported to the microwave sensor is calculated using continuous wave phase difference ranging or frequency-modulated continuous wave ranging. By filtering out environmental noise through Fast Fourier Transform, the characteristic frequencies reflected by the items to be transported are extracted, and real-time dimensions and positioning data of the items to be transported are obtained. And / or, The infrared thermal imager and infrared ranging sensor are used to acquire infrared outlines of the items to be transported, and distance data from the left and right edges of the items to the center line of the conveyor line or a preset position, including: On the same side as the high-definition camera, an infrared thermal imager is installed above the transmission line. Multiple infrared thermal imagers are used to form a multi-view mode to provide additional parallax information. The ranging is switched by binocular combination, and infrared data of the items to be transported is obtained by temperature difference recognition. Based on the infrared ranging sensor emitting near-infrared rays, the distance to the infrared object to be transported is detected by triangulation or time-of-flight principle. Using an uncooled infrared focal plane array, a preset infrared band is detected, and the temperature difference between the item to be transported and the environment is captured to generate a thermal image. The region to be transported is segmented based on the region growing algorithm. The infrared outline of the item to be transported is extracted based on the infrared item to be transported data, and the distance data from the left and right edges of the item to be transported to the center line of the transport line or the preset positioning is calculated.

3. The method for dynamic positioning of items on a conveyor line based on multimodal ranging according to claim 1, characterized in that, The process of determining the positioning direction of the item to be transported based on the fused location data and the preset positioning position includes: When comparing the size detection values ​​of the various items to be transported from the data, if the data deviation is greater than the preset size deviation value, the sensor data with the highest weight is selected as the size detection value of the item to be transported. Using the centerline of the conveyor line as a reference, the deviation between the center of the item to be conveyed and the centerline of the conveyor line or the preset positioning is calculated by fusing the edge coordinates of the laser, the geometric center of vision, and the edge distance data of microwave or infrared. Based on dual-sided multi-point detection using laser sensors, combined with the Hough transform results from vision, the angle between the plane of the item to be transported and the conveyor line is calculated. Based on image information captured by a high-definition camera, the system outputs the dimensions of the cargo that extend beyond the edge of the items to be transported and identifies information about fragile items.

4. The method for dynamic positioning of items on a conveyor line based on multimodal ranging according to claim 3, characterized in that, The power adjustment unit includes a servo motor, a push rod, a first flexible contact pad, and a position encoder; the process of driving the power adjustment unit to push the item to be transported to the preset positioning position according to the positioning direction of the item to be transported includes: The servo motor drives the push rod, which, in conjunction with the first flexible contact pad, pushes the item to be transported to the preset positioning position in real time according to the calculated offset. The position encoder is built into the push rod and is used to provide feedback on the push rod's stroke. And / or, The power adjustment unit further includes a baffle motor, a baffle, and a second flexible contact pad; the process of driving the power adjustment unit to push the item to be transported to the preset positioning position according to the positioning direction of the item to be transported includes: The baffle motor controls the lifting and lowering of the baffle, which, together with the second flexible contact pad, ensures that the front end of the item to be conveyed is flush with the front end. When the item to be transported stops, the baffle is mechanically stopped to eliminate the effect of tilting; and / or, The process of driving the power adjustment unit to push the item to be transported to the preset positioning position according to the positioning direction of the item to be transported also includes: Based on the fused core parameters, the distance that the push rod on one side needs to move is calculated for the preset positioning offset, so as to obtain the basic offset correction displacement. If the item to be transported has a tilt angle, the tilt compensation amount is calculated based on the tilt angle and the length of the item to be transported, and the tilt side push rod is pushed forward by the tilt compensation amount to eliminate the tilt through the differential movement of the push rods on both sides. If the goods are detected to be beyond the edge of the items to be transported, the movement distance of the side push rod beyond the edge is reduced accordingly, so that the other side push rod can bear the main adjustment amount. And / or, The process of driving the power adjustment unit to push the item to be transported to the preset positioning position according to the positioning direction of the item to be transported, and the subsequent process, also includes: The calculated target displacement and adjustment speed are converted into control signals for the servo motor or baffle motor to drive the push rod or baffle to move. The pressure threshold of the push rod or baffle is output synchronously to monitor overload risk in real time; The system tracks the location of items to be transported in real time, forming a closed-loop control for adjusting the location of the items to be transported, thus ensuring the final positioning effect.

5. A dynamic positioning system for conveyor items based on multimodal ranging, characterized in that, A method for performing dynamic positioning of conveyor items based on multimodal ranging as described in any one of claims 1-4, comprising: The multi-source detection unit, in response to the start of the conveyor line, detects the items to be conveyed on the conveyor line to obtain various data on the items to be conveyed; The data fusion unit performs multi-source fusion processing on the data of the various items to be transported according to preset weights, so as to achieve compatible identification of items to be transported of various specifications and materials, and obtain the location fusion data of the items to be transported. The positioning calculation unit determines the positioning direction of the item to be transported based on the fused location data of the item to be transported and the preset positioning location; The control and monitoring unit, based on the positioning direction of the item to be transported, drives the power adjustment unit to push the item to be transported to the preset positioning position.

6. A dynamic positioning system for conveyor items based on multimodal ranging according to claim 5, characterized in that, The multi-source detection unit includes a laser sensor; the laser sensors are symmetrically installed on both sides of the conveyor line, with two sets installed symmetrically on each side at a distance of 300mm from left to right.

7. A dynamic positioning system for conveyor items based on multimodal ranging according to claim 6, characterized in that, The multi-source detection unit includes a high-definition camera; the high-definition camera is installed at a height of 2.5m and tilted at 45° above the conveyor line, forming a three-lens high-definition vision detection system.

8. A dynamic positioning system for conveyor items based on multimodal ranging according to claim 7, characterized in that, The multi-source detection unit includes an infrared thermal imager; the infrared thermal imager and the high-definition camera are mounted on the same side above the conveyor line.

9. A dynamic positioning system for conveyor items based on multimodal ranging according to claim 8, characterized in that, The multi-source detection unit includes a microwave sensor; the microwave sensor is installed at the bottom of both sides of the conveyor line with a beam angle of 30° and is parallel to the conveyor line.

Images