AGV pallet fork positioning method and device based on distance measurement sensing analysis and medium

Abstract

The invention discloses an AGV pallet fork positioning method and device based on distance measurement sensing analysis and a medium, and relates to the technical field of data processing, and the method comprises the steps: configuring a distance measurement sensor, a vision sensor and an orientation sensor on a target AGV, and constructing a sensing collection device group; activating the equipment group, and collecting multi-dimensional sensing data of the target AGV, including distance, image and azimuth data; carrying out environment position positioning by using the image data, and traversing a three-dimensional simulation space of a target scene to extract a local reference space; determining a first relative position of the target AGV pallet fork in a local reference space in combination with the multi-dimensional sensing data; and analyzing the local reference space to obtain the second relative position of the target object, comparing the difference between the two positions to obtain positioning difference information, and optimizing target AGV pallet fork positioning based on the difference information, thereby achieving the technical effects of improving the pallet fork positioning efficiency and the operation safety.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and in particular to an AGV fork positioning method, equipment and medium based on distance measurement sensor analysis. Background Art

[0002] In the docking of the new energy back-end (side fork AGV for picking up and placing materials on shelves), the number of storage locations is about 1,000 to 5,000. Due to the left and right deviations in the walking process of the AGV, there is a relative deviation when the side fork AGV picks up and places materials and the materials fall on the guide block, causing the materials to occasionally fall on the fork guide block. This can easily cause the AGV to protrude from the vehicle body and cause secondary scratches on the equipment or uneven materials on the shelves. There are technical problems such as low fork positioning accuracy, low positioning efficiency, and impact on operation safety. Summary of the invention

[0003] The present invention provides an AGV fork positioning method, device and medium based on ranging sensor analysis to solve the technical problems of low fork positioning accuracy, low positioning efficiency and impact on operation safety in the prior art, and achieve the technical effect of improving fork positioning efficiency and operation safety.

[0004] In a first aspect, the present invention provides an AGV fork positioning method based on ranging sensor analysis, wherein the method comprises:

[0005] Set up distance measuring sensors, visual sensors and orientation sensors on the target AGV, and configure and generate a sensor acquisition device group.

[0006] The sensor acquisition device group is activated to collect multi-dimensional ranging sensor data of the target AGV, wherein the multi-dimensional ranging sensor data includes distance data, image data and azimuth angle data.

[0007] The environment position is located based on the image data, and according to the environment position location result, the three-dimensional simulation space of the target scene is traversed to extract the local reference space.

[0008] The distance data, the image data and the azimuth data are combined to determine a first relative spatial position of the fork of the target AGV in the local reference space.

[0009] The local reference space is parsed to obtain a second relative spatial position of the target object, and the first relative spatial position is compared with the second relative spatial position to obtain difference positioning information.

[0010] The fork positioning of the target AGV is optimized according to the differential positioning information.

[0011] In a feasible method, a distance measuring sensor, a visual sensor and an orientation sensor are set on the target AGV, and a sensor acquisition device group is configured to be generated, including:

[0012] Determine sensor specifications based on the positioning requirements of the target scene.

[0013] According to the sensor specification information, the interactive target AGV performs matching screening on existing sensors and defines a sensor deployment list according to the matching screening results.

[0014] The sensor layout of the target AGV is updated using the sensor layout list, and the distance measuring sensor, the visual sensor and the orientation sensor are connected to the sensor acquisition sink to obtain the sensor acquisition device group.

[0015] In a feasible manner, the environment position is located based on the image data, and according to the environment position location result, the three-dimensional simulation space of the target scene is traversed to extract the local reference space, including:

[0016] The image data is subjected to image denoising and image enhancement, and key feature information in the image is extracted.

[0017] Environmental feature recognition based on the key feature information is performed using a computer vision method to generate an environmental feature vector.

[0018] The environmental feature vector database of the target scene is traversed for matching, and the environmental position positioning result of the target AGV is determined.

[0019] The local reference space is obtained by combining the environmental position positioning result with the preset space extraction constraint and performing local space extraction in the three-dimensional simulation space.

[0020] In a feasible manner, combining the distance data, the image data and the azimuth data to determine the first relative spatial position of the fork of the target AGV in the local reference space includes:

[0021] The azimuth angle data is used as the first registration basis, and the first-level registration is performed in combination with the initial spatial orientation coordinate system of the target AGV.

[0022] The distance data is used as a second registration basis, and the position of the first-level registration result is corrected based on the triangulation method to obtain a second-level registration result.

[0023] The image data is analyzed to extract contour information and texture information of the target object as auxiliary registration basis, and the secondary registration result is verified by the auxiliary registration basis.

[0024] If the verification is passed, the first relative spatial position is determined based on the secondary registration result and the relative position of the target AGV fork within the machine.

[0025] In a feasible manner, the target object is the operation target object of the target AGV fork, including a transport target object and a placement target object, and the second relative spatial position is the relative spatial position of the operation task control point corresponding to the target object in the local reference space.

[0026] In one feasible embodiment, the method further includes:

[0027] The target AGV obtains the fork operation record, and screens and determines the non-empty operation area from the fork operation record, and outputs it as a feedback monitoring area set.

[0028] The feedback monitoring area set is continuously monitored by the scene monitoring device of the target scene, and feedback correction is performed on the target object according to the continuous monitoring result.

[0029] In one feasible embodiment, the method further includes:

[0030] The number of feedback corrections is accumulated through the accumulator.

[0031] If the feedback correction times exceed the first preset times or the continuous feedback correction times exceed the second preset times, the decision correction for fork positioning optimization is performed according to the feedback correction record, wherein the first preset times is greater than the second preset times.

[0032] In one feasible embodiment, the method further includes:

[0033] Get the safety fence of the target space.

[0034] The relative distance between the target AGV fork and the safety fence is monitored in real time.

[0035] The relative distance is weighted and corrected according to the safety fence level and the real-time task level of the target AGV. If the correction result is less than a preset safety threshold, the target AGV is automatically stopped.

[0036] In a second aspect, the present invention further provides an electronic device, comprising: a memory for storing executable instructions; and a processor for implementing the AGV fork positioning method based on ranging sensor analysis provided by the present invention when executing the executable instructions stored in the memory.

[0037] In a third aspect, the present invention further provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the AGV fork positioning method based on distance measurement sensor analysis provided by the present invention.

[0038] The present invention discloses an AGV fork positioning method, device and medium based on distance sensing analysis, comprising: setting a distance sensor, a visual sensor and an azimuth sensor on a target AGV, and configuring and generating a sensor acquisition device group; activating the sensor acquisition device group to collect multi-dimensional distance sensing data of the target AGV, wherein the multi-dimensional distance sensing data includes distance data, image data and azimuth data; performing environmental position positioning based on the image data, and traversing a three-dimensional simulation space of a target scene according to the environmental position positioning result to extract a local reference space; determining a first relative spatial position of a fork of the target AGV in the local reference space by combining the distance data, the image data and the azimuth data; parsing the local reference space to obtain a second relative spatial position of the target object, and comparing the first relative spatial position with the second relative spatial position to obtain differential positioning information; optimizing the positioning of the fork of the target AGV according to the differential positioning information. The AGV fork positioning method, device and medium based on distance sensing analysis disclosed by the present invention solve the technical problems of low fork positioning accuracy, low positioning efficiency and impact on operation safety, and achieve the technical effect of improving fork positioning efficiency and operation safety. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] Figure 1 It is a schematic diagram of the flow of the AGV fork positioning method based on distance measurement sensor analysis of the present invention;

[0040] Figure 2 It is a schematic diagram of the structure of an exemplary electronic device of the present invention.

[0041] Description of the reference numerals: processor 31 , memory 32 , input device 33 , output device 34 . DETAILED DESCRIPTION

[0042] The above technical solution will be described in detail below in conjunction with the accompanying drawings and specific implementation methods of the specification to better understand the above technical solution. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all of the embodiments of the present invention. It should be understood that the present invention is not limited to the example embodiments used only to explain the present invention. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention. In addition, it should be noted that, for the convenience of description, only the parts related to the present invention are shown in the drawings, rather than all of them.

[0043] Embodiment 1

[0044] Figure 1 The figure is a flow chart of the AGV fork positioning method based on distance sensing analysis of the present invention, wherein the method comprises:

[0045] Set up distance measuring sensors, visual sensors and orientation sensors on the target AGV, and configure and generate a sensor acquisition device group.

[0046] Specifically, the distance sensor is used to measure the distance between the AGV and surrounding obstacles or target points, and is mainly used for obstacle avoidance, navigation and path planning, including laser radar (LiDAR), ultrasonic sensors, and infrared sensors. Visual sensors are used for environmental perception, target recognition, path tracking, and understanding of complex scenes, including RGB cameras, depth cameras, panoramic cameras, etc. Azimuth sensors are used to obtain the posture, direction, and position information of the AGV to support navigation and path planning. Available ones include IMU (inertial measurement unit), electronic compass, GPS module, UWB (ultra-wideband) module, etc.

[0047] Specifically, the ranging sensor, visual sensor and orientation sensor are integrated into the AGV to form a complete sensor acquisition device group, including a data processing unit (such as an embedded main control board) corresponding to the acquisition device group composed of the ranging sensor, visual sensor and orientation sensor. Through the above configuration, the sensor acquisition device group of the AGV can achieve efficient data collection and environmental perception to meet the task execution requirements in complex scenarios.

[0048] In some embodiments, a distance measuring sensor, a visual sensor, and an orientation sensor are set on the target AGV, and a sensor acquisition device group is configured to be generated, including:

[0049] Determine sensor specification information based on the positioning requirement information of the target scene;

[0050] According to the sensor specification information, the interactive target AGV performs matching screening on the existing sensors, and defines a sensor deployment list according to the matching screening results;

[0051] Update the sensor layout of the target AGV with the sensor layout list, and connect the distance sensor, visual sensor and orientation sensor to the sensor acquisition sink to obtain the sensor acquisition device group.

[0052] Specifically, first, according to the positioning requirement information of the target scene, the required sensor types and performance indicators are analyzed, including ranging range, accuracy, response time, etc., to determine the sensor specification information that meets these requirements, such as the performance parameters of laser radar (LIDAR), camera, gyroscope, etc.; then, check the existing sensors on the target AGV for matching screening, determine the available sensor models and specifications, and define the sensor deployment list that needs to be added or replaced based on the matching screening results, where the sensor deployment list includes sensor models, quantities and installation locations.

[0053] Specifically, according to the sensor deployment list, the physical deployment of sensors is carried out on the target AGV, including fixing, wiring and interface connection, so as to connect the ranging sensors, visual sensors and orientation sensors to the sensor acquisition sink through appropriate interfaces. The sensor acquisition sink is a centralized data processing unit or control module. Then, the software system of the AGV is updated, including the sensor driver and configuration parameters, and the sensor acquisition equipment group is functionally tested to ensure the accuracy and real-time nature of the sensor data.

[0054] Through the above process, it can be ensured that the target AGV is equipped with sensors suitable for the target scene and can accurately perform positioning and navigation.

[0055] The sensor acquisition device group is activated to collect multi-dimensional ranging sensor data of the target AGV, wherein the multi-dimensional ranging sensor data includes distance data, image data and azimuth angle data.

[0056] Specifically, all sensors installed on the AGV are started, including ranging sensors (such as lidar or ultrasonic sensors), visual sensors (such as cameras), and orientation sensors (such as gyroscopes or compasses), and the sensor system is initialized to ensure that all sensors work synchronously and are adjusted to the correct working mode, that is, according to the operating requirements and processing capabilities of the AGV, a suitable data acquisition sampling rate is set to ensure data consistency and availability; then, the ranging sensor is used to collect real-time distance data between the AGV and surrounding obstacles, and the visual sensor (such as a camera) is used to capture image data within the AGV's field of view for environmental recognition and navigation. The current azimuth angle data of the AGV is obtained through the orientation sensor (such as IMU, gyroscope, or compass) to determine the direction of the AGV.

[0057] The environment position is located based on the image data, and according to the environment position location result, the three-dimensional simulation space of the target scene is traversed to extract the local reference space.

[0058] Specifically, environmental position positioning refers to the use of environmental data collected by sensors combined with known environmental information (such as maps or three-dimensional models) to determine the current position and direction of the AGV in the environment, such as which shelf the AGV is on in the environment and its current direction.

[0059] Specifically, the three-dimensional simulation space of the target scene refers to the three-dimensional digital representation of the real scene, a virtual three-dimensional model containing all the geometric information of the target scene, which can reflect the structure and layout of the actual environment and is used for simulation, analysis and planning; the local reference space refers to a small three-dimensional area related to the current device position extracted from the three-dimensional simulation space of the target scene. It is a local subset of the environment and is used for real-time perception, decision-making and operation of AGV.

[0060] In some embodiments, performing environmental location positioning based on the image data, and traversing the three-dimensional simulation space of the target scene according to the environmental location positioning result to extract the local reference space includes:

[0061] Performing image denoising and image enhancement on the image data, and extracting key feature information in the image;

[0062] Performing environmental feature recognition based on the key feature information by using a computer vision method to generate an environmental feature vector;

[0063] Traversing the environmental feature vector database of the target scene for matching, and determining the environmental position positioning result of the target AGV;

[0064] The local reference space is obtained by combining the environmental position positioning result with the preset space extraction constraint and performing local space extraction in the three-dimensional simulation space.

[0065] Specifically, image preprocessing is first performed, and a denoising algorithm is applied to the collected image data to reduce the noise in the image and improve the accuracy of subsequent processing; the contrast and brightness of the image are enhanced to make the key features in the image more obvious; then, key feature information in the image, such as edges, corners, textures, etc., is extracted through feature engineering.

[0066] Specifically, computer vision technology is used to analyze key feature information extracted from the image, identify and describe the characteristics of the environment, and illustratively include matching in the target image based on a known template image to find the location of the target object; using deep learning target detection methods (such as YOLO, Faster R-CNN) to detect specific objects or areas in the environment; using a deep convolutional neural network (CNN) to classify the image at the pixel level and extract semantic information of different areas of the environment (such as roads, buildings, plants, etc.); then, converting the environmental features in the image into a numerical vector, that is, generating an environmental feature vector as input for subsequent calculations or modeling.

[0067] Specifically, the environmental feature vector database of the target scene is traversed, and the generated environmental feature vector is matched with the vector in the database to determine the environmental position positioning result of the AGV, such as the row and column position of the shelf; then, the environmental position positioning result and the preset spatial extraction constraints (such as distance, angle, height, etc.) are combined to determine the local spatial area that needs to be extracted in the three-dimensional simulation space to obtain the local reference space.

[0068] Through this process, image data is used to accurately locate the environmental position, and the local reference space related to its position is extracted in the three-dimensional simulation space, which improves the accuracy and environmental adaptability of the AGV and enables it to operate safely and efficiently in complex environments.

[0069] The distance data, the image data and the azimuth data are combined to determine a first relative spatial position of the fork of the target AGV in the local reference space.

[0070] In some embodiments, combining the distance data, the image data, and the azimuth data to determine a first relative spatial position of a fork of the target AGV in the local reference space includes:

[0071] Taking the azimuth angle data as the first registration basis, a first-level registration is performed in combination with the initial spatial orientation coordinate system of the target AGV;

[0072] Using the distance data as a second registration basis, performing position correction on the first registration result based on triangulation to obtain a second registration result;

[0073] Parsing the image data, extracting contour information and texture information of the target object as auxiliary registration basis, and verifying the secondary registration result by using the auxiliary registration basis;

[0074] If the verification is passed, the first relative spatial position is determined based on the secondary registration result and the relative position of the target AGV fork within the machine.

[0075] Specifically, first, the azimuth data is used as the first alignment basis, and the first-level alignment (i.e., coarse alignment) is performed in combination with the initial spatial orientation coordinate system of the AGV. For example, a rotation matrix or quaternion is used to transform the fork position from the AGV coordinate system to the local reference space, and the initial orientation of the AGV relative to the local reference space is determined, wherein the initial spatial orientation coordinate system is used for the preliminary position and orientation of the AGV in the local reference space.

[0076] Specifically, based on the primary registration result, the distance data is used as the basis for the second registration, and the triangulation method is used to calculate the precise position of the AGV fork relative to the known fixed point to obtain the secondary registration result.

[0077] Furthermore, the image data is parsed to extract the contour information and texture information of the target object, and the extracted contour information and texture information are used as the basis for auxiliary registration; then, according to the secondary registration result, the extracted image features are compared with the known feature templates in the local reference space to verify the accuracy of the secondary registration result. If the position error between the spatial position of the image feature and the position of the secondary registration result is less than a preset threshold, the verification is judged to have passed.

[0078] Optionally, if the verification fails, the secondary registration result is further refined using image features, illustratively, by fine-tuning using an optimization algorithm (such as an ICP algorithm) to ensure high-precision matching of position and orientation.

[0079] Specifically, after verification, the secondary registration result is integrated with the in-machine relative position parameters of the target AGV fork, so as to calculate the final position of the fork in the local reference space.

[0080] Through the above process, the fork position of the AGV can be accurately determined in the local reference space. The primary and secondary registrations provide position information based on sensor data, while image data analysis provides visual verification to ensure the accuracy of position data. This method improves the operating accuracy of the AGV, helps reduce operating errors, and enhances the adaptability and reliability of the AGV in complex environments.

[0081] The local reference space is parsed to obtain a second relative spatial position of the target object, and the first relative spatial position is compared with the second relative spatial position to obtain difference positioning information.

[0082] Specifically, the data in the local reference space, including preset markers, known object positions or environmental features, are analyzed to determine the position of the target object in the local reference space, that is, to calculate the second relative spatial position of the target object relative to the AGV fork.

[0083] Specifically, the first relative spatial position (the position of the AGV fork) is compared with the second relative spatial position (the position of the target object), and the difference between the two positions, including distance difference, angle difference, etc., is calculated to obtain difference positioning information, which is used to adjust the position and orientation of the AGV fork to ensure accurate alignment or transportation of the target object.

[0084] In some embodiments, the target object is an operation target object of a target AGV fork, including a transport target object and a placement target object, and the second relative spatial position is a relative spatial position of an operation task control point corresponding to the target object in the local reference space.

[0085] Specifically, the target object refers specifically to the operating target object of the AGV fork, including the target object that needs to be transported and the target object to be placed (such as a shelf), and the second relative spatial position refers to the key control points in the operating task, such as the relative positions of the grasping points, placement points or other key operating points in the local reference space.

[0086] The fork positioning of the target AGV is optimized according to the differential positioning information.

[0087] Specifically, the positioning optimization of the target AGV fork is performed based on the differential positioning information to ensure that the AGV can accurately perform handling and placement tasks. Optionally, the fork positioning optimization includes adjusting the fork of the target AGV to a preset working position (ideal working position) to compensate for position deviations and optimize the trajectory of the target AGV's fork operation through path planning methods, thereby improving operating efficiency and accuracy and ensuring cargo safety.

[0088] Through the above process, the positioning accuracy of the AGV's fork is improved, the success rate and efficiency of handling and placement tasks are increased, and the optimized positioning strategy enables the AGV to better adapt to the complex and changing working environment, reduce operational errors, and improve the reliability and stability of the automated logistics system.

[0089] In some embodiments, the method further comprises:

[0090] Recording the target AGV to obtain the fork operation record, and screening and determining the non-empty operation area from the fork operation record, and outputting it as a feedback monitoring area set;

[0091] The feedback monitoring area set is continuously monitored by the scene monitoring device of the target scene, and feedback correction is performed on the target object according to the continuous monitoring result.

[0092] Specifically, the fork operation record includes the fork's operation time, location coordinates, target object information (such as the type, weight, and size of the goods), and related operation status (such as picking up and releasing goods); the non-empty operation area is defined as the coordinates or area range that contains clear target object interactions, that is, the area where goods exist after the operation, which requires continuous monitoring.

[0093] Specifically, real-time data collection is performed on the non-empty work area set through fixed or mobile monitoring equipment in the scene (such as cameras, lidars). The monitoring data content includes the status of target objects in the environment (whether they have been moved or damaged) and dynamic change information of the area (such as whether there are new obstacles).

[0094] Furthermore, the target object status is compared and analyzed to identify abnormal situations (such as the target object is placed incorrectly, the target is lost, the target slips, collapses, etc.), and corresponding corrective measures are taken, such as re-moving it to the correct location, calling spare monitoring equipment or mobile equipment to search for the target again; using AGV forks to make slight adjustments to correct the slip status; suspending work tasks in related areas, initiating manual intervention or collaborative robots for cleaning and recovery, etc.

[0095] Through comprehensive identification and accurate processing of abnormal conditions, the error rate in logistics or warehousing operations can be effectively reduced, the security of target object storage and management can be improved, and at the same time, operational efficiency and the intelligence level of the automation system can be optimized.

[0096] In some implementations, the method further includes:

[0097] The number of feedback corrections is accumulated through an accumulator;

[0098] If the feedback correction times exceed the first preset times or the continuous feedback correction times exceed the second preset times, the decision correction for fork positioning optimization is performed according to the feedback correction record, wherein the first preset times is greater than the second preset times.

[0099] Specifically, an accumulator is set, and each time the fork positioning optimization is performed, the number of feedback corrections is accumulated in the accumulator; then, the number of corrections in the accumulator is monitored to compare whether it exceeds a first preset number or whether the continuous feedback corrections exceed a second preset number, wherein the first preset number is greater than the second preset number.

[0100] Specifically, if the number of corrections in the accumulator exceeds the first preset number, or the number of continuous feedback corrections exceeds the second preset number, the decision correction process is triggered, including adjusting the positioning optimization parameters, such as navigation parameters, control algorithm parameters, etc. according to the feedback correction records; improving the positioning optimization strategy, such as introducing a new algorithm, optimizing the path planning logic, etc.; and then, updating the adjusted parameters and optimization strategy to the AGV control system to achieve more accurate fork positioning.

[0101] Through the above method, the positioning process of the AGV fork can be systematically monitored and optimized. By setting the preset number of times and making decision corrections based on the feedback correction records, unnecessary corrections can be reduced and the stability and accuracy of positioning can be improved.

[0102] In some embodiments, the method further comprises:

[0103] Get the security fence of the target space;

[0104] Real-time monitoring of the relative distance between the target AGV fork and the safety fence;

[0105] The relative distance is weighted and corrected according to the safety fence level and the real-time task level of the target AGV. If the correction result is less than a preset safety threshold, the target AGV is automatically stopped.

[0106] Specifically, the safety fence can be physical (such as actual existing fences and obstacles, generated by sensors scanning the environment in real time) or virtual (through safety boundaries extracted from predefined maps or scene models), representing the areas that the AGV needs to avoid colliding or approaching during operation.

[0107] Specifically, safety fences have different fence levels, which indicate the safety importance or degree of danger of the fence. For example, low-level ordinary areas do not require special protection; high-level dangerous areas (such as next to high-temperature equipment, the edge of a deep pit, etc.) require higher security.

[0108] Specifically, the real-time position and posture of the fork are obtained through sensors (such as encoders, IMUs, lidars, and vision systems), and then the minimum distance between the fork and the safety fence is calculated based on the real-time position of the fork and the position of the safety fence.

[0109] Specifically, the task level indicates the importance or urgency of the AGV's current task. According to the safety fence level and the task level, the relative distance can be weightedly corrected. If the corrected distance is less than the safety threshold, it is considered that there is a safety risk and measures need to be taken, such as emergency stop of the target AGV. The safety threshold indicates the minimum allowable distance between the fork and the safety fence.

[0110] By combining the safety fence level and task level, the relative distance between the AGV and the fence is weighted and corrected, and the dynamic adjustment mechanism of the AGV is automatically stopped when a safety risk is detected. This improves operational safety while flexibly optimizing operational efficiency according to task requirements.

[0111] In summary, the AGV fork positioning method based on distance measurement sensor analysis provided by the present invention has the following technical effects:

[0112] The method is implemented by setting a distance measuring sensor, a visual sensor and an azimuth sensor on the target AGV, and configuring and generating a sensor acquisition device group; activating the sensor acquisition device group to collect multi-dimensional distance measuring sensor data of the target AGV, wherein the multi-dimensional distance measuring sensor data includes distance data, image data and azimuth data; performing environmental position positioning based on the image data, and traversing the three-dimensional simulation space of the target scene according to the environmental position positioning result to extract the local reference space; combining the distance data, image data and azimuth data to determine the first relative spatial position of the fork of the target AGV in the local reference space; parsing the local reference space to obtain the second relative spatial position of the target object, and comparing the first relative spatial position with the second relative spatial position to obtain differential positioning information; optimizing the positioning of the fork of the target AGV according to the differential positioning information, thereby achieving the technical effect of improving the positioning efficiency of the fork and the safety of the operation.

[0113] Embodiment 2

[0114] Figure 2 The schematic structural diagram of an exemplary electronic device provided for the present invention shows a block diagram of an exemplary electronic device suitable for implementing an embodiment of the present invention. Figure 2The electronic device shown is only an example and should not limit the functions and scope of use of the embodiments of the present invention. Figure 2 As shown, the electronic device includes a processor 31, a memory 32, an input device 33 and an output device 34; the number of processors 31 in the electronic device can be one or more. Figure 2 Taking a processor 31 as an example, the processor 31, memory 32, input device 33 and output device 34 in the electronic device can be connected through a bus or other means. Figure 2 The example of connecting through bus is taken in the following.

[0115] The memory 32 is a computer-readable storage medium that can be used to store software programs, computer executable programs and modules, such as program instructions / modules corresponding to the AGV fork positioning method based on distance sensing analysis in the embodiment of the present invention. The processor 31 executes various functional applications and data processing of the computer device by running the software programs, instructions and modules stored in the memory 32, that is, realizing the above-mentioned AGV fork positioning method based on distance sensing analysis.

[0116] Embodiment 3

[0117] The present embodiment provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the following steps are implemented: a distance measuring sensor, a visual sensor and an azimuth sensor are set on a target AGV, and a sensor acquisition device group is configured to be generated; the sensor acquisition device group is activated to collect multi-dimensional distance measuring sensor data of the target AGV, wherein the multi-dimensional distance measuring sensor data includes distance data, image data and azimuth angle data; environmental position positioning is performed based on the image data, and according to the environmental position positioning result, a three-dimensional simulation space of a target scene is traversed to extract a local reference space; a first relative spatial position of a fork of the target AGV in the local reference space is determined by combining the distance data, the image data and the azimuth angle data; the local reference space is parsed to obtain a second relative spatial position of the target object, and the first relative spatial position is compared with the second relative spatial position to obtain difference positioning information; and the fork positioning of the target AGV is optimized according to the difference positioning information.

[0118] It should be noted that in the above embodiments, the description of each embodiment has its own emphasis, and for parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0119] It should be understood that the embodiments disclosed in the present invention and the above description can enable those skilled in the art to use the present invention to implement the present invention. At the same time, the present invention is not limited to the above-mentioned embodiments. It should be understood that those skilled in the art can still modify the technical solutions recorded in the above-mentioned embodiments, or replace some of the technical features therein by equivalents; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should be included in the protection scope of the present invention.

Claims

1. The AGV fork positioning method based on distance sensing analysis is characterized by: The method comprises: Set up distance measuring sensors, visual sensors and orientation sensors on the target AGV, and configure and generate a sensor acquisition device group; Activate the sensor collection device group to collect multi-dimensional ranging sensor data of the target AGV, wherein the multi-dimensional ranging sensor data includes distance data, image data and azimuth data; Performing environmental position positioning based on the image data, and traversing the three-dimensional simulation space of the target scene according to the environmental position positioning result to extract the local reference space; Determine a first relative spatial position of a fork of a target AGV in the local reference space by combining the distance data, the image data and the azimuth data; Parsing the local reference space to obtain a second relative spatial position of the target object, and comparing the first relative spatial position with the second relative spatial position to obtain difference positioning information; The fork positioning of the target AGV is optimized according to the differential positioning information.

2. The AGV fork positioning method based on distance sensing analysis as claimed in claim 1, characterized in that: Set up distance sensors, visual sensors, and orientation sensors on the target AGV, and configure and generate a sensor acquisition device group, including: Determine sensor specification information based on the positioning requirement information of the target scene; According to the sensor specification information, the interactive target AGV performs matching screening on the existing sensors, and defines a sensor deployment list according to the matching screening results; The sensor layout of the target AGV is updated using the sensor layout list, and the distance measuring sensor, the visual sensor and the orientation sensor are connected to the sensor acquisition sink to obtain the sensor acquisition device group.

3. The AGV fork positioning method based on distance sensing analysis as claimed in claim 2, characterized in that: Performing environmental position positioning based on the image data, and traversing the three-dimensional simulation space of the target scene according to the environmental position positioning result to extract the local reference space, including: Performing image denoising and image enhancement on the image data, and extracting key feature information in the image; Performing environmental feature recognition based on the key feature information by using a computer vision method to generate an environmental feature vector; Traversing the environmental feature vector database of the target scene for matching, and determining the environmental position positioning result of the target AGV; The local reference space is obtained by combining the environmental position positioning result with the preset space extraction constraint and performing local space extraction in the three-dimensional simulation space.

4. The AGV fork positioning method based on distance sensing analysis as claimed in claim 3, characterized in that: Combining the distance data, the image data and the azimuth data, determining a first relative spatial position of a fork of the target AGV in the local reference space includes: Taking the azimuth angle data as the first registration basis, a first-level registration is performed in combination with the initial spatial orientation coordinate system of the target AGV; Using the distance data as a second registration basis, performing position correction on the first registration result based on triangulation to obtain a second registration result; Parsing the image data, extracting contour information and texture information of the target object as auxiliary registration basis, and verifying the secondary registration result by using the auxiliary registration basis; If the verification is passed, the first relative spatial position is determined based on the secondary registration result and the relative position of the target AGV fork within the machine.

5. The AGV fork positioning method based on distance sensing analysis as claimed in claim 4, characterized in that: The target object is the operation target object of the target AGV fork, including a transport target object and a placement target object, and the second relative spatial position is the relative spatial position of the operation task control point corresponding to the target object in the local reference space.

6. The AGV fork positioning method based on distance sensing analysis as claimed in claim 1, characterized in that: The method further comprises: Recording the target AGV to obtain the fork operation record, and screening and determining the non-empty operation area from the fork operation record, and outputting it as a feedback monitoring area set; The feedback monitoring area set is continuously monitored by the scene monitoring device of the target scene, and feedback correction is performed on the target object according to the continuous monitoring result.

7. The AGV fork positioning method based on distance sensing analysis as claimed in claim 6, characterized in that: The method further comprises: The number of feedback corrections is accumulated through an accumulator; If the feedback correction times exceed the first preset times or the continuous feedback correction times exceed the second preset times, the decision correction for fork positioning optimization is performed according to the feedback correction record, wherein the first preset times is greater than the second preset times.

8. The AGV fork positioning method based on distance sensing analysis as claimed in claim 1, characterized in that: The method further comprises: Get the security fence of the target space; Real-time monitoring of the relative distance between the target AGV fork and the safety fence; The relative distance is weighted and corrected according to the safety fence level and the real-time task level of the target AGV. If the correction result is less than a preset safety threshold, the target AGV is automatically stopped.

9. An electronic device, characterized in that: The electronic device comprises: A memory for storing executable instructions; The processor is used to implement the AGV fork positioning method based on ranging sensor analysis as described in any one of claims 1 to 8 when executing the executable instructions stored in the memory.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the AGV fork positioning method based on distance measurement sensor analysis as described in any one of claims 1 to 8 is implemented.

Images