Port heavy-load AGV intelligent navigation method and device

Abstract

The invention provides an intelligent navigation method and device for a port heavy-load AGV (Automatic Guided Vehicle). The port heavy-load AGV intelligent navigation method comprises the following steps: acquiring surrounding environment real-time information of an AGV and a task-level instruction message corresponding to the AGV; determining positioning information and global map information of the AGV based on the surrounding environment real-time information, wherein the global map information comprises a global static map and dynamic obstacle information; trajectory planning is carried out based on the task-level instruction message, the positioning information and the global map information, and a planned trajectory corresponding to the AGV is generated; and controlling the AGV to operate according to the planned trajectory. According to the port heavy-load AGV intelligent navigation method provided by the embodiment of the invention, the operation efficiency of the terminal AGV can be effectively improved.

Description

Technical Field

[0001] The present invention relates to the technical field of port transportation, and in particular to a method and device for intelligent navigation of a heavy-load AGV in a port. Background Art

[0002] As an important hub for sea and land logistics, container terminals have become the main trend in the construction of modern container terminals in the context of large-scale ships. With the increasing requirements for operating efficiency, reducing labor costs, terminal environmental protection and energy saving, automated terminals have become the general direction of industry development with their significant advantages such as high safety and reliability, high operating efficiency, high site utilization, environmental friendliness and low labor costs.

[0003] Therefore, how to use automation and intelligent technology to further improve the operational efficiency of terminal AGV has become an urgent problem to be solved. Summary of the invention

[0004] In view of this, the present invention provides a port heavy-load AGV intelligent navigation method and device, which can effectively optimize the motion trajectory and motion control strategy of the AGV and improve the operating efficiency of the AGV.

[0005] In order to solve the above technical problems, the present invention adopts the following technical solutions:

[0006] The port heavy-load AGV intelligent navigation method according to the first embodiment of the present invention includes:

[0007] Obtaining real-time information about the surrounding environment of the AGV and task-level instruction messages corresponding to the AGV;

[0008] Determine the positioning information and global map information of the AGV based on the real-time information of the surrounding environment, wherein the global map information includes a global static map and dynamic obstacle information;

[0009] Perform trajectory planning based on the task-level instruction message, the positioning information, and the global map information to generate a planned trajectory corresponding to the AGV;

[0010] The AGV is controlled to run according to the planned trajectory.

[0011] Further, the determining the positioning information of the AGV based on the real-time information of the surrounding environment includes:

[0012] Performing data calibration on the real-time information of the surrounding environment to obtain corresponding calibration parameters;

[0013] Based on the real-time information of the surrounding environment and its corresponding calibration parameters, the positioning information of the AGV is determined using a preset positioning algorithm.

[0014] Further, the determining of global map information based on the surrounding environment real-time information includes:

[0015] Building a global static map based on the real-time information of the surrounding environment, and determining dynamic obstacle information in the surrounding environment of the AGV;

[0016] The global static map is updated according to the dynamic obstacle information to obtain the global map information.

[0017] Furthermore, the building of a global static map based on the real-time information of the surrounding environment and determining the dynamic obstacle information in the surrounding environment of the AGV includes:

[0018] Performing data calibration on the real-time information of the surrounding environment to obtain corresponding calibration parameters;

[0019] Based on the real-time information of the surrounding environment and its corresponding calibration parameters, the global static map is constructed using a preset map construction method;

[0020] Based on the real-time information of the surrounding environment and its corresponding calibration parameters, the dynamic obstacle information in the surrounding environment of the AGV is determined using a preset perception algorithm.

[0021] Further, performing trajectory planning based on the task-level instruction message, the positioning information, and the global map information to generate a planned trajectory corresponding to the AGV includes:

[0022] Determine the target position of the AGV according to the task-level instruction message;

[0023] Based on the target position and the positioning information, a preset trajectory planning algorithm is used, combined with the global static map and the dynamic obstacle information, to generate a planned trajectory corresponding to the AGV.

[0024] Further, based on the target position and the positioning information, using a preset trajectory planning algorithm, combining the global static map and the dynamic obstacle information, generating a planned trajectory corresponding to the AGV includes:

[0025] Based on the target position and the positioning information, a global trajectory planning algorithm is used to generate a planning trajectory corresponding to the AGV in combination with the global static map and the dynamic obstacle information;

[0026] The planned trajectory is adjusted in real time based on the dynamic obstacle information to obtain an adjusted planned trajectory.

[0027] Further, controlling the AGV to run according to the planned trajectory includes:

[0028] Perform kinematic and dynamic calculations on the AGV according to the planned trajectory, the task-level instruction message, and the dynamic obstacle information to determine the action decision of the AGV;

[0029] A control decision instruction is generated according to the action decision, and the operation of the AGV is controlled based on the control decision instruction.

[0030] Furthermore, the real-time information of the surrounding environment includes port facility information and / or obstacle information.

[0031] The port heavy-load AGV intelligent navigation device according to the second embodiment of the present invention includes:

[0032] A data acquisition module is used to obtain real-time information about the surrounding environment of the AGV and task-level instruction messages corresponding to the AGV;

[0033] A positioning and map building module, used to determine the positioning information and global map information of the AGV based on the real-time information of the surrounding environment, wherein the global map information includes a global static map and dynamic obstacle information;

[0034] A path planning module, used to perform trajectory planning based on the task-level instruction message, the positioning information and the global map information, and generate a planned trajectory corresponding to the AGV;

[0035] A control module is used to control the operation of the AGV according to the planned trajectory.

[0036] Furthermore, the positioning and map building module includes:

[0037] A positioning and mapping module, used to determine the positioning information of the AGV based on the real-time information of the surrounding environment and to construct a global static map;

[0038] A dynamic perception module, used to determine dynamic obstacle information in the surrounding environment of the AGV based on the real-time information of the surrounding environment;

[0039] The electronic map module is used to construct the global map information based on the global static map and the dynamic obstacle information.

[0040] The above technical solution of the present invention has at least one of the following beneficial effects:

[0041] According to the intelligent navigation method for heavy-duty AGVs in ports of the present invention, real-time information about the surrounding environment of the AGV is acquired to determine the positioning information and global map information of the AGV, and trajectory planning is performed based on the acquired task-level instruction messages, positioning information, and global map information, thereby generating a planned trajectory that matches the AGV in real time, and finally controlling the AGV to reach the predetermined destination based on the planned trajectory. Thus, various static / dynamic obstacles in the route can be effectively identified, the optimization of the AGV route can be ensured, and the operating efficiency of the AGV on the port terminal can be effectively improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] Figure 1 This is a flow chart of a port heavy-load AGV intelligent navigation method according to an embodiment of the present invention;

[0043] Figure 2 It is a flow chart of a port heavy-load AGV intelligent navigation method according to another embodiment of the present invention;

[0044] Figure 3 It is a front view of the arrangement of sensors on the AGV in the intelligent navigation method for heavy-load AGV in a port according to an embodiment of the present invention;

[0045] Figure 4 A top view of the arrangement of sensors on the AGV in the port heavy-load AGV intelligent navigation method according to an embodiment of the present invention;

[0046] Figure 5 It is a left view of the arrangement of sensors on the AGV in the intelligent navigation method for heavy-load AGVs in ports according to an embodiment of the present invention;

[0047] Figure 6 This is a module schematic diagram of a port heavy-load AGV intelligent navigation device according to an embodiment of the present invention;

[0048] Figure 7 This is a module schematic diagram of a port heavy-load AGV intelligent navigation device according to another embodiment of the present invention;

[0049] Figure 8 Schematic diagram of software control of the port heavy-load AGV intelligent navigation device according to an embodiment of the present invention.

[0050] Figure numerals: 1. Main laser radar; 2. Side blind spot laser radar; 3. Top positioning laser radar; 4. GPS positioning system; 5. Inertial measurement unit IMU; 6. Industrial camera. DETAILED DESCRIPTION

[0051] In order to make the purpose, technical solution and advantages of the embodiment of the present invention clearer, the technical solution of the embodiment of the present invention will be clearly and completely described below in conjunction with the drawings of the embodiment of the present invention. Obviously, the described embodiment is a part of the embodiment of the present invention, not all of the embodiments. Based on the described embodiment of the present invention, all other embodiments obtained by ordinary technicians in this field belong to the scope of protection of the present invention.

[0052] Unless otherwise defined, the technical terms or scientific terms used in the present invention shall have the usual meanings understood by persons with ordinary skills in the field to which the present invention belongs. The words "first", "second" and similar words used in the present invention do not indicate any order, quantity or importance, but are only used to distinguish different components. Words such as "connected" or "connected" are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. "Up", "down", "left", "right" and the like are only used to indicate relative positional relationships. When the absolute position of the object being described changes, the relative positional relationship also changes accordingly.

[0053] Below we first combine the attached Figure 1 The intelligent navigation method for a heavy-load AGV in a port according to an embodiment of the present invention is described in detail. The method can be applied to a controller of an intelligent navigation system for a heavy-load AGV in a port.

[0054] According to the first aspect of the present invention, the port heavy-load AGV intelligent navigation method is as follows: Figure 1 As shown, it may include:

[0055] S1: Obtain the real-time information of the surrounding environment of the AGV and the task-level instruction message corresponding to the AGV.

[0056] In an embodiment of the present invention, the real-time information of the surrounding environment may include port facility information and / or obstacle information. The port facility information may be, for example, facility information of a quay crane, a column, an artificial island, etc.; the obstacle information may include static obstacle information and dynamic obstacle information, both of which are obtained through sensors on the AGV. The task-level instruction message may be a control instruction issued by the software platform, and the task performed by the AGV and its corresponding target position may be determined according to the control instruction.

[0057] As an example, the arrangement of the sensors is as follows Figure 3 , Figure 4 , Figure 5, a 32-line main laser radar 1 is respectively arranged at the two corners of a diagonal line of the AGV, with a horizontal upward scanning range of 0-10° and a horizontal downward scanning range of 0-16°, and can effectively detect a target with a reflectivity of 10% at a distance of 110 meters. On both sides of the length direction of the AGV, a lateral blind laser radar 2 is respectively arranged. The two lateral blind laser radars are located at both ends of the AGV, which can effectively detect a target with a reflectivity of 10% at a distance of 180 meters, and the horizontal field of view is 120°. An upper positioning laser radar 3 is respectively arranged at both ends of the length direction of the upper surface of the AGV, which can effectively detect a target with a reflectivity of 10% at a distance of 30 meters, with a horizontal field of view of 360° and a vertical field of view of 90°. Industrial cameras 6 are respectively arranged on both sides of the width direction of the AGV. A GPS positioning system 4 is also respectively arranged on the upper surface of the AGV near the 32-line laser radar. An inertial measurement unit IMU5 (also known as a posture sensor) and a controller with AI computing power are set inside the AGV.

[0058] S2: Determine the positioning information and global map information of the AGV based on the real-time information of the surrounding environment.

[0059] In the embodiment of the present invention, the positioning information and global map information of the AGV can be determined by obtaining real-time information of the surrounding environment such as port facility information and / or obstacle information. The global map information includes a global static map and dynamic obstacle information, that is, the global map information is generated by combining the global static map with the dynamic obstacle information.

[0060] It can be understood that by updating the global static map with real-time dynamic obstacle information to construct real-time global map information, the accuracy of the map information can be guaranteed, and the efficiency and reliability of port transportation can be guaranteed.

[0061] S3: Perform trajectory planning based on task-level command messages, positioning information, and global map information to generate a planned trajectory corresponding to the AGV.

[0062] In the embodiment of the present invention, by receiving task-level instruction messages, real-time trajectory route calculation and planning can be performed based on the positioning information of the AGV and real-time global map information, and finally the most efficient planned trajectory matching the AGV to reach the predetermined destination is generated.

[0063] S4: Control the AGV operation according to the planned trajectory.

[0064] In the embodiment of the present invention, the AGV can be controlled to run according to its predetermined trajectory by obtaining the calculated planned trajectory, which is accurate and reliable.

[0065] In summary, according to the intelligent navigation method for heavy-duty AGVs in ports according to the embodiment of the present invention, the real-time information of the surrounding environment of the AGV is acquired to determine the positioning information and global map information of the AGV, and the trajectory is planned according to the obtained task-level instruction message, positioning information and global map information, so as to generate a planned trajectory corresponding to the AGV in real time, and finally control the AGV to reach the predetermined destination according to the planned trajectory. In this way, various static / dynamic obstacles in the route can be effectively identified, the optimization of the AGV route can be ensured, and the operating efficiency of the AGV on the port terminal can be effectively improved.

[0066] In some embodiments, determining the positioning information of the AGV based on the real-time information of the surrounding environment may include:

[0067] Perform data calibration on the real-time information of the surrounding environment to obtain the corresponding calibration parameters;

[0068] Based on the real-time information of the surrounding environment and its corresponding calibration parameters, the preset positioning algorithm is used to determine the positioning information of the AGV.

[0069] In the embodiment of the present invention, the existing data calibration method can be used to pre-calibrate the received real-time information of the surrounding environment to obtain calibration parameters, and then the AGV is positioned using a preset positioning algorithm based on the received real-time information of the surrounding environment and the calibration parameters to obtain the positioning information of the AGV. Among them, the preset positioning algorithm can be pre-set according to actual needs, for example, high-precision positioning data of the AGV can be obtained through multi-level loose coupling fusion of multiple positioning algorithms, and the embodiment of the present invention does not specifically limit this.

[0070] Specifically, in the loosely coupled algorithm, the independent operation of the Global Navigation Satellite System (GNSS) module and the Inertial Navigation System (INS) module is utilized. The GNSS module outputs the real-time kinematic (RTK) positioning result at a low frequency, and the INS module outputs the inertial measurement result at a high frequency, and then the data of both are input into the data processing module. When the RTK positioning result is not received, the filter infers the current positioning data based on the measurement result of the INS module. When the RTK positioning result is received, the filter compares the difference between the calculation results of the RTK and INS modules, establishes an error model, estimates the accumulated error of the INS module, and feeds back the error compensation to the INS module, and outputs the best estimate of the positioning data. After receiving the error compensation, the INS module corrects the accumulated error and repeats the cycle. The advantage of using this method of fusion is that it is easy to implement and has relatively stable performance.

[0071] In some embodiments, determining global map information based on surrounding environment real-time information may include:

[0072] Build a global static map based on real-time information about the surrounding environment and determine dynamic obstacle information in the AGV's surrounding environment;

[0073] The global static map is updated according to the dynamic obstacle information to obtain global map information.

[0074] In the embodiment of the present invention, on the basis of the acquired global static map, real-time dynamic obstacle information can be updated to finally obtain real-time global map information, which can further ensure the real-time and accuracy of the map information, thereby improving the efficiency of AGV navigation operation.

[0075] That is to say, a global static map is pre-built by acquiring real-time information about the surrounding environment, and then the dynamic layer of the global static map is updated in real time by acquiring dynamic obstacle information in the AGV's surrounding environment (that is, real-time perception of the AGV's surrounding environment), and global map information is formed based on the global static map updated by the dynamic layer.

[0076] Specifically, the global static map provides a fixed reference frame, which is used to display the basic components of the working environment around the AGV, such as fixed static environmental factors such as lane lines and fixed obstacles. The dynamic layer is used to present dynamically changing environmental factors in real time, such as moving vehicles, pedestrians, and temporary obstacles. By receiving and processing the perceived data information (through means such as object detection and object tracking), these data will be processed and superimposed on the dynamic layer of the global static map to form a comprehensive real-time environmental model (i.e., global map information). This method allows the AGV to effectively perceive and make decisions in a dynamic environment, thereby improving the accuracy of navigation and path planning. The real-time updated dynamic layer can not only display the immediate changes in the current environment, but also predict potential dynamic obstacles through intelligent algorithms, thereby providing the latest information support for the AGV's path planning.

[0077] In some embodiments, constructing a global static map based on real-time information of the surrounding environment and determining dynamic obstacle information in the surrounding environment of the AGV may include:

[0078] Perform data calibration on the real-time information of the surrounding environment to obtain the corresponding calibration parameters;

[0079] Based on the real-time information of the surrounding environment and its corresponding calibration parameters, a global static map is constructed using a preset map construction method;

[0080] Based on the real-time information of the surrounding environment and its corresponding calibration parameters, the preset perception algorithm is used to determine the dynamic obstacle information in the AGV's surrounding environment.

[0081] In the embodiment of the present invention, after obtaining the real-time information of the surrounding environment and its corresponding calibration parameters, a preset map construction method can be used to construct a map to obtain a global static map. The preset map construction method can be preset according to actual needs, for example, a high-precision global static map can be constructed by a laser SLAM algorithm and RTK-GNSS information, which is not specifically limited in the embodiment of the present invention.

[0082] In an embodiment of the present invention, after obtaining the real-time information of the surrounding environment and its corresponding calibration parameters, the preset perception algorithm can be used to perceive the dynamic obstacle information. The preset perception algorithm can be preset according to actual needs. For example, the laser perception algorithm and the visual perception algorithm can be used to obtain high-precision perception information of the environment, that is, the dynamic obstacle information in the surrounding environment of the AGV. After that, the perception information (such as the information of facilities such as quay crane columns and artificial islands, obstacle information, etc.) can be updated in real time in the global static map, thereby obtaining global map information.

[0083] In some embodiments, trajectory planning is performed based on task-level instruction messages, positioning information, and global map information, and generating a planned trajectory corresponding to the AGV may include:

[0084] Determine the target position of the AGV based on the task-level instruction message;

[0085] Based on the target position and positioning information, the preset trajectory planning algorithm is used, combined with the global static map and dynamic obstacle information, to generate the corresponding planning trajectory of the AGV.

[0086] In the embodiment of the present invention, after receiving the task-level instruction message, the target location information to be reached by the AGV is obtained, and then the target location information to be reached by the AGV, the positioning information of the AGV itself, the global static map and the dynamic obstacle information are combined, and the preset trajectory planning algorithm is used to calculate the planned trajectory of the AGV to reach the corresponding target location. Among them, the preset trajectory planning algorithm can be pre-set according to actual needs, and the embodiment of the present invention does not specifically limit this.

[0087] That is to say, the information of the target position to be reached by the AGV is obtained according to the instructions issued by the task-level instruction message, and trajectory planning is performed based on the target position to be reached by the AGV and the real-time positioning information. In addition, the global static map and dynamic obstacle information are combined to generate the optimal planned walking trajectory corresponding to the AGV.

[0088] In some embodiments, based on the target position and positioning information, using a preset trajectory planning algorithm, combined with the global static map and the dynamic obstacle information, generating a planned trajectory corresponding to the AGV includes:

[0089] Based on the target position and positioning information, the global trajectory planning algorithm is used to combine the global static map and dynamic obstacle information to generate the planning trajectory corresponding to the AGV;

[0090] The planned trajectory is adjusted in real time based on the dynamic obstacle information to obtain the adjusted planned trajectory.

[0091] That is to say, when performing trajectory planning calculations, it includes two parts: global trajectory calculation and local trajectory calculation. The global trajectory planning algorithm can be used to calculate the overall route of the AGV's planned travel, that is, to generate the planned trajectory corresponding to the AGV. Then, the trajectory of the corresponding local area in the trajectory is planned in real time according to the dynamic obstacle information (for example, by sensing the size of the obstacle and controlling the running speed to avoid obstacles more accurately), and finally the adjusted planned trajectory is obtained, thereby ensuring the optimal planned trajectory.

[0092] In some embodiments, Figure 2 As shown, controlling the AGV operation according to the planned trajectory may include:

[0093] S41: Perform kinematic and dynamic calculations on the AGV according to the planned trajectory, task-level command messages, and dynamic obstacle information to determine the action decision of the AGV;

[0094] S42: Generate a control decision instruction according to the action decision, and control the operation of the AGV based on the control decision instruction.

[0095] In an embodiment of the present invention, kinematic and dynamic calculations can be performed on the AGV based on the planned trajectory, task-level instruction messages, and dynamic obstacle information, thereby obtaining corresponding matching AGV action decisions, which can also include work optimization decisions, trapped release decisions, automatic obstacle avoidance decisions, emergency action decisions, etc. The operation of the AGV is controlled based on the real-time response action decisions obtained after the calculation, which can maximize the guarantee of high-efficiency operation of the AGV.

[0096] That is to say, after receiving the task-level instruction message, that is, issuing the control instruction, obstacle avoidance information perception (i.e. based on dynamic obstacle information) must be implemented on the basis of the generated planning trajectory, and the final AGV action decision is obtained by combining the calculation. In other words, after generating the planning trajectory, it is necessary to combine the acquisition of real-time dynamic information to make the final intelligent decision to ensure that the AGV can make the best action decision based on real-time data. For example, when other AGVs or pedestrians are detected near the AGV, the motion trajectory and behavior patterns of these dynamic objects are analyzed, and the motion trajectory of the AGV is adjusted based on this information. Through real-time analysis of perception data, the motion control strategy of the AGV can be optimized, the risk of collision can be reduced, transportation efficiency can be improved, and autonomous obstacle avoidance function can be realized, so that the AGV can quickly take avoidance measures when facing unexpected obstacles, thereby ensuring safe and smooth operation.

[0097] The following describes in detail a port heavy-load AGV intelligent navigation device according to an embodiment of the second aspect of the present invention.

[0098] In some embodiments, the port heavy-load AGV intelligent navigation device 100 according to the second aspect of the present invention is as follows: Figure 6 As shown, it may include:

[0099] The data acquisition module 110 is used to acquire the real-time information of the surrounding environment of the AGV and the task-level instruction message corresponding to the AGV;

[0100] A positioning and map building module 120, for determining the positioning information and global map information of the AGV based on the real-time information of the surrounding environment, wherein the global map information includes a global static map and dynamic obstacle information;

[0101] A path planning module 130, configured to perform trajectory planning based on the task-level instruction message, the positioning information, and the global map information, and generate a planned trajectory corresponding to the AGV;

[0102] The control module 140 is used to control the operation of the AGV according to the planned trajectory.

[0103] In other embodiments, Figure 7 As shown, the positioning and map building module 120 may include:

[0104] A positioning and mapping module 121 is used to determine the positioning information of the AGV based on the real-time information of the surrounding environment and to construct a global static map;

[0105] A dynamic perception module 122, configured to determine dynamic obstacle information in the surrounding environment of the AGV based on the real-time information of the surrounding environment;

[0106] The electronic map module 133 is used to construct the global map information based on the global static map and the dynamic obstacle information.

[0107] Specifically, Figure 8 As shown, first, the multi-sensor hardware system on the heavy-duty AGV (such as RTK combined navigation system, IMU system, lidar system, and industrial camera, etc.) collects environmental information such as quay crane facilities, columns, and artificial islands, and transmits the sensor data to the controller of the AGV.

[0108] In the controller's software system, the automatic calibration module first calibrates each sensor data to obtain the calibration parameters of each sensor, and transmits the calibration parameters to the high-precision positioning and mapping module and the high-precision dynamic perception module. Among them, the high-precision positioning and mapping module accepts sensor data and calibration parameters, obtains high-precision positioning data of the AGV and builds a static map. The high-precision dynamic perception module receives sensor data and calibration parameters to obtain high-precision perception information of the environment.

[0109] The high-precision electronic map module maintains the current AGV's environmental map information locally. The high-precision electronic map module is used to perform real-time dynamic map updates on static maps (i.e., update real-time perception information, such as information on facilities such as quay crane columns and artificial islands, obstacle information, etc.) to obtain global map information. In addition, the high-precision electronic map module also communicates with the upper-level software platform (host computer) in real time to exchange positioning, perception, and map information.

[0110] The path planning module receives task-level command messages from the upper-level platform software and performs real-time trajectory planning based on global map information and high-precision positioning data.

[0111] The intelligent decision control module receives the task-level command message from the upper platform software, and intelligently generates the control decision of the AGV based on real-time trajectory planning and obstacle avoidance information from the high-precision dynamic perception module. Among them, the intelligent decision module usually contains advanced algorithms and models, such as path planning algorithms, obstacle avoidance algorithms, and behavior prediction models, which can process a large amount of dynamic information and make quick decisions.

[0112] It should be noted here that, based on the trajectory planning output by the path planning module, the intelligent decision-making control module further perceives obstacle avoidance information (i.e. dynamic obstacle information) through a high-precision dynamic perception module, and based on the obstacle avoidance information, transmits the perceived surrounding obstacle information to the intelligent decision-making control module in real time, thereby realizing autonomous decision-making of the AGV.

[0113] The above is a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A port heavy-load AGV intelligent navigation method, characterized in that: include: Obtaining real-time information about the surrounding environment of the AGV and task-level instruction messages corresponding to the AGV; Determine the positioning information and global map information of the AGV based on the real-time information of the surrounding environment, wherein the global map information includes a global static map and dynamic obstacle information; Perform trajectory planning based on the task-level instruction message, the positioning information, and the global map information to generate a planned trajectory corresponding to the AGV; The AGV is controlled to run according to the planned trajectory.

2. The port heavy-load AGV intelligent navigation method according to claim 1 is characterized in that: The determining of the positioning information of the AGV based on the real-time information of the surrounding environment includes: Performing data calibration on the real-time information of the surrounding environment to obtain corresponding calibration parameters; Based on the real-time information of the surrounding environment and its corresponding calibration parameters, the positioning information of the AGV is determined using a preset positioning algorithm.

3. The port heavy-load AGV intelligent navigation method according to claim 1 is characterized in that: Determining the global map information based on the surrounding environment real-time information includes: Building a global static map based on the real-time information of the surrounding environment, and determining dynamic obstacle information in the surrounding environment of the AGV; The global static map is updated according to the dynamic obstacle information to obtain the global map information.

4. The port heavy-load AGV intelligent navigation method according to claim 3 is characterized in that: The building of a global static map based on the real-time information of the surrounding environment and determining the dynamic obstacle information in the surrounding environment of the AGV includes: Performing data calibration on the real-time information of the surrounding environment to obtain corresponding calibration parameters; Based on the real-time information of the surrounding environment and its corresponding calibration parameters, the global static map is constructed using a preset map construction method; Based on the real-time information of the surrounding environment and its corresponding calibration parameters, the dynamic obstacle information in the surrounding environment of the AGV is determined using a preset perception algorithm.

5. The port heavy-load AGV intelligent navigation method according to claim 1 is characterized in that: The performing trajectory planning based on the task-level instruction message, the positioning information and the global map information to generate the planned trajectory corresponding to the AGV includes: Determine the target position of the AGV according to the task-level instruction message; Based on the target position and the positioning information, a preset trajectory planning algorithm is used, combined with the global static map and the dynamic obstacle information, to generate a planned trajectory corresponding to the AGV.

6. The port heavy-load AGV intelligent navigation method according to claim 5 is characterized in that: The generating of the planned trajectory corresponding to the AGV based on the target position and the positioning information, using a preset trajectory planning algorithm, and combining the global static map and the dynamic obstacle information includes: Based on the target position and the positioning information, a global trajectory planning algorithm is used to generate a planning trajectory corresponding to the AGV in combination with the global static map and the dynamic obstacle information; The planned trajectory is adjusted in real time based on the dynamic obstacle information to obtain an adjusted planned trajectory.

7. The port heavy-load AGV intelligent navigation method according to claim 1 is characterized in that: Controlling the AGV to run according to the planned trajectory includes: Perform kinematic and dynamic calculations on the AGV according to the planned trajectory, the task-level instruction message, and the dynamic obstacle information to determine the action decision of the AGV; A control decision instruction is generated according to the action decision, and the operation of the AGV is controlled based on the control decision instruction.

8. The port heavy-load AGV intelligent navigation method according to claim 1 is characterized in that: The real-time information of the surrounding environment includes port facility information and / or obstacle information.

9. A port heavy-load AGV intelligent navigation device, characterized in that: include: A data acquisition module is used to obtain real-time information about the surrounding environment of the AGV and task-level instruction messages corresponding to the AGV; A positioning and map building module, used to determine the positioning information and global map information of the AGV based on the real-time information of the surrounding environment, wherein the global map information includes a global static map and dynamic obstacle information; A path planning module, used to perform trajectory planning based on the task-level instruction message, the positioning information and the global map information, and generate a planned trajectory corresponding to the AGV; A control module is used to control the operation of the AGV according to the planned trajectory.

10. The port heavy-load AGV intelligent navigation device according to claim 9, characterized in that: The positioning and map building module includes: A positioning and mapping module, used to determine the positioning information of the AGV based on the real-time information of the surrounding environment and to construct a global static map; A dynamic perception module, used to determine dynamic obstacle information in the surrounding environment of the AGV based on the real-time information of the surrounding environment; The electronic map module is used to construct the global map information based on the global static map and the dynamic obstacle information.

Images