Dynamic path planning method for warehouse-in and warehouse-out of workbin robot
By extracting key turning points and performing sliding sampling on the inbound and outbound paths of the bin robot, and dynamically updating the path, the problems of insufficient path optimization and untimely response to moving obstacles in the existing technology are solved, achieving more efficient and safer path planning and improving warehouse operation efficiency.
Patent Information
- Application Number
- CN202411054081.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-02
- Publication Date
- 2024-11-15
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
Existing dynamic path planning methods for bin robots in the process of entering and leaving the warehouse have shortcomings. They are not precise enough in handling turning points, resulting in suboptimal paths, increased travel time and energy consumption. At the same time, they lack a rapid response mechanism for moving obstacles, which can easily lead to collisions or frequent stops, affecting warehouse operation efficiency.
By reading the turning points on the global path, key turning points are extracted to form a sequence of sub-target points. Sliding sampling is performed based on a sliding window to determine changes in obstacles, adjust dynamic description features, and add moving obstacles as new sub-target nodes. The path is then dynamically updated to generate a dynamic path that conforms to the actual environment.
The optimized path at turns reduced unnecessary turns and travel distances, lowered robot travel time and energy consumption, improved response speed to moving obstacles, avoided collisions and waiting times, and enhanced warehouse operational efficiency.
Smart Images

Figure CN118963357A_ABST
Abstract
Description
Technical Field
[0001] The embodiments of the present application relate to the field of digital twins, and in particular to a dynamic path planning method for a container robot to enter and exit a warehouse. Background Art
[0002] In the field of automated warehousing and logistics, container robots are widely used in the handling, storage and management of goods, significantly improving the operational efficiency of warehouses. In order to ensure that container robots can efficiently and safely complete the tasks of entering and exiting the warehouse in a complex warehouse environment, path planning technology becomes the key. Traditional path planning methods are mainly based on static environments, that is, when planning paths, it is assumed that the positions of all obstacles in the warehouse are fixed. However, in actual applications, the warehouse environment is often dynamically changing. For example, the movement of other robots or staff, temporarily stacked goods, etc., may become moving obstacles.
[0003] Although the existing dynamic path planning methods take into account the mobility of obstacles in the environment, they often have the following problems: First, the turning points on the global path are not handled finely enough, resulting in the generated dynamic path being not optimized enough, which increases the robot's driving time and energy consumption; second, there is a lack of a fast and effective response mechanism for the sudden appearance of moving obstacles, which can easily cause the robot to collide with obstacles or require frequent stops and waiting, affecting the overall operational efficiency of the warehouse. Summary of the invention
[0004] The purpose of this application is to propose a dynamic path planning method for a material box robot to enter and exit the warehouse, so as to solve or overcome the above-mentioned technical problems existing in the prior art.
[0005] According to a first aspect of an embodiment of the present application, a dynamic path planning method for a container robot entering and exiting a warehouse is provided, which includes:
[0006] Read all turning points on the global path planned by the container robot for inbound and outbound storage;
[0007] Extracting a number of key turning points from all the turning points, wherein one of the key turning points is used as a sub-target point, so as to form a sub-target point sequence according to the order of the key turning points on the corresponding global path;
[0008] Sliding sampling is performed on the sub-target point sequence based on a set sliding window to obtain a sampling target point group;
[0009] Determine whether a moving obstacle appears near the sampling target point group. If so, adjust the dynamic description feature of the material box robot and use the moving obstacle as a newly added sub-target node;
[0010] Based on all the newly added sub-target points determined, the global path is dynamically updated to generate a dynamic path for the container robot to enter and exit the warehouse.
[0011] Optionally, the step of reading all turning points on the global path of the storage and retrieval planning of the material box robot includes:
[0012] Reading a global environment model of the warehouse where the material box robot is located, wherein the global environment model defines an obstacle area and a free running area;
[0013] Parsing the in-and-out instructions of the material box robot to determine the current starting position and the current target point position of the material box robot;
[0014] According to the starting point position, the current target point position and the global environment model, a global path search is performed for the material box robot to enter and exit the warehouse, and a plurality of optional global paths are obtained to form an optional global path set;
[0015] Based on the set global path constraints, the optional global path set is solved to select an optional global path that meets the set path constraints and uses it as the planned global path.
[0016] Optionally, the step of reading all turning points on the global path of the storage and retrieval planning of the material box robot includes:
[0017] Extract all adjacent path nodes on the global path;
[0018] Direction analysis is performed on each of the adjacent path nodes to read all turning points on the global path of the material box robot's in-and-out storage planning.
[0019] Optionally, extracting all adjacent path nodes on the global path includes:
[0020] A coordinate point trajectory on a global path, based on which all adjacent path nodes on the global path are determined and extracted.
[0021] Optionally, the performing of direction analysis on each of the adjacent path nodes to read all turning points on the global path of the storage-in and storage-out planning of the material box robot includes:
[0022] Performing direction analysis on each of the adjacent path nodes to obtain a direction vector;
[0023] Based on the tolerance of the turning radius setting of the material box robot, the dot product and the modulus between two adjacent direction vectors are calculated to calculate the angle between the two adjacent direction vectors;
[0024] According to the angle, all turning points on the global path of the material box robot's storage and retrieval planning are determined and read.
[0025] Optionally, extracting a plurality of key turning points from all the turning points, wherein one of the key turning points is used as a sub-target point, so as to form a sub-target point sequence according to the order of the key turning points on the corresponding global path, comprises:
[0026] The key turning point screening description is read to extract a number of key turning points from all the turning points based on the key turning point screening description, wherein one key turning point is used as a sub-target point to form a sub-target point sequence according to the sequence of the key turning points on the corresponding global path.
[0027] Optionally, the reading of the key turning point screening description to extract a plurality of key turning points from all the turning points based on the key turning point screening description, wherein one of the key turning points is used as a sub-target point to form a sub-target point sequence according to the sequence of the key turning points on the corresponding global path, includes:
[0028] Based on the set turning angle threshold, the distance threshold of adjacent turning points, and the path curvature threshold of the turn, several turning points whose turning angle is greater than the turning angle threshold, the distance between adjacent turning points is greater than the distance threshold, and the path curvature of the turn is greater than the path curvature threshold are extracted from all the turning points as several key turning points, one of the key turning points is used as a sub-target point to form a sub-target point sequence according to the sequence of the key turning points on the corresponding global path, one of the key turning points is used as a sub-target point to form a sub-target point sequence according to the sequence of the key turning points on the corresponding global path.
[0029] Optionally, the forming of the sub-target point sequence according to the sequence of key turning points on the corresponding global path, wherein one key turning point is used as a sub-target point, and the forming of the sub-target point sequence according to the sequence of key turning points on the corresponding global path includes:
[0030] According to the sequence of the key turning points on the corresponding global path, each of the key turning points is numbered to obtain a path node;
[0031] Based on the size of the serial number, the path nodes are sorted, and the key turning points are added to the initialized target point sequence in sequence according to the sequence of the key turning points on the corresponding global path to form the sub-target point sequence.
[0032] Optionally, performing sliding sampling on the sub-target point sequence based on a set sliding window to obtain a sampled target point group includes:
[0033] A speed sampling space constructed according to the moving speed limit and dynamic constraints of the material box robot is obtained, so as to perform sliding sampling on the sub-target point sequence based on a set sliding window to obtain a sampling target point group.
[0034] Optionally, the acquiring of a speed sampling space constructed according to a moving speed limit and a dynamic constraint of the material box robot, so as to obtain a sampling target point group by sliding sampling the sub-target point sequence based on a set sliding window, comprises:
[0035] A speed sampling space is obtained based on the moving speed limit of the material box robot and the dynamic constraint, wherein the moving speed limit includes the maximum and minimum values of the speed and acceleration, and the dynamic constraint is the maximum braking distance and the minimum braking distance when the material box robot encounters an obstacle.
[0036] According to the speed sampling space, the speed combination of the material box robot is predicted, so as to obtain a sampling target point group by sliding sampling of the sub-target point sequence based on a set sliding window.
[0037] Optionally, predicting the speed combination of the material box robot according to the speed sampling space, so as to obtain a sampling target point group by sliding sampling the sub-target point sequence based on a set sliding window, comprises:
[0038] According to the speed sampling space, the speed combination of the material box robot is predicted, and the speed combination is mapped to the set sliding window to determine from the sub-target point sequence that the material box robot can reach the sub-target point or the vicinity of the sub-target point within the time of the sliding window, and the sub-target point that is reached or the sub-target point that satisfies the vicinity is used as the sampling target point to form a sampling target point group.
[0039] Optionally, the determining whether a moving obstacle appears near the sampling target point group, and if so, adjusting the dynamic description feature of the material box robot, and using the moving obstacle as a newly added sub-target node, includes:
[0040] The material box robot is subjected to environmental perception when moving along a static path to determine dynamic obstacles with a set distance similarity to the sampling target point group. If such obstacles appear, it is determined that a moving obstacle appears near the sampling target point group. If such obstacles appear, at least one of the moving speed and the moving direction of the material box robot is adjusted, and the moving obstacle is added as a new sub-target node. The dynamic description features of the material box robot include at least one of the moving speed and the moving direction.
[0041] Optionally, the adjusting at least one of the moving speed and the moving direction of the material box robot and taking the moving obstacle as a newly added sub-target node includes:
[0042] Based on the set evaluation function, the current dynamic characteristics of the material box robot are evaluated to obtain a dynamic evaluation value, so as to adjust at least one of the moving speed and the travel direction of the material box robot according to the dynamic evaluation value, and the moving obstacle is used as a new sub-target node. The evaluation function includes: a direction angle evaluation sub-function, which represents the direction angle deviation of the moving obstacle relative to the current target point at the current speed; a distance evaluation sub-function, which represents the distance deviation of the moving obstacle relative to the current target point at the current speed; and a current speed magnitude evaluation sub-function, which represents the magnitude of the current speed.
[0043] Optionally, dynamically updating the global path based on all the newly added sub-target points determined to generate a dynamic path for the container robot to enter and exit the warehouse includes:
[0044] According to the degree of proximity between each newly added sub-target point and the current target point position of the material box robot, the free running position points that the material box robot can reach to avoid the newly added sub-target points are determined from the free running area, so that the global path is dynamically updated based on the free running position points to generate a dynamic path for the material box robot to enter and exit the warehouse.
[0045] In the present invention, all turning points on the global path of the in-and-out planning of the material box robot are first read, and several key turning points are extracted therefrom. The selection of these key turning points is based on their importance and sequence on the global path, thereby ensuring the optimization of the dynamic path at the turning point. By forming a sub-target point sequence and sliding sampling the sub-target point sequence based on a set sliding window, the path planning process is further refined. This method enables the dynamic path to be more in line with the actual environment, reduces unnecessary turns and driving distances, thereby reducing the driving time and energy consumption of the robot. In the path planning process, the influence of moving obstacles is particularly considered. By judging whether there are moving obstacles near the sampling target point group, and adjusting the dynamic description characteristics of the material box robot in time, the robot can quickly adapt to changes in the environment. The moving obstacles are used as new sub-target nodes, and the global path is dynamically updated based on all the new sub-target points, generating a dynamic path that is more in line with the actual situation. This method effectively avoids the situation where the robot collides with obstacles or frequently stops and waits, and improves the overall operational efficiency of the warehouse. BRIEF DESCRIPTION OF THE DRAWINGS
[0046] Hereinafter, some specific embodiments of the present application will be described in detail in an exemplary and non-limiting manner with reference to the accompanying drawings. The same reference numerals in the accompanying drawings indicate the same or similar components or parts. It should be understood by those skilled in the art that these drawings are not necessarily drawn to scale. In the accompanying drawings:
[0047] Figure 1 A schematic flow chart of a dynamic path planning method for a material box robot entering and exiting a warehouse provided in an embodiment of the present application. DETAILED DESCRIPTION
[0048] In order to enable those skilled in the art to better understand the technical solutions in the embodiments of the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments in the embodiments of the present application should fall within the scope of protection of the embodiments of the present application.
[0049] Figure 1 A schematic diagram of a dynamic path planning method for a material box robot entering and exiting a warehouse provided in an embodiment of the present application. Figure 1 As shown, it includes:
[0050] Read all turning points on the global path planned by the container robot for inbound and outbound storage;
[0051] Extracting a number of key turning points from all the turning points, wherein one of the key turning points is used as a sub-target point, so as to form a sub-target point sequence according to the order of the key turning points on the corresponding global path;
[0052] Sliding sampling is performed on the sub-target point sequence based on a set sliding window to obtain a sampling target point group;
[0053] Determine whether a moving obstacle appears near the sampling target point group. If so, adjust the dynamic description feature of the material box robot and use the moving obstacle as a newly added sub-target node;
[0054] Based on all the newly added sub-target points determined, the global path is dynamically updated to generate a dynamic path for the container robot to enter and exit the warehouse.
[0055] In the present invention, all turning points on the global path of the in-and-out planning of the material box robot are first read, and several key turning points are extracted therefrom. The selection of these key turning points is based on their importance and sequence on the global path, thereby ensuring the optimization of the dynamic path at the turning point. By forming a sub-target point sequence and sliding sampling the sub-target point sequence based on a set sliding window, the path planning process is further refined. This method enables the dynamic path to be more in line with the actual environment, reduces unnecessary turns and driving distances, thereby reducing the driving time and energy consumption of the robot. In the path planning process, the influence of moving obstacles is particularly considered. By judging whether there are moving obstacles near the sampling target point group, and adjusting the dynamic description characteristics of the material box robot in time, the robot can quickly adapt to changes in the environment. The moving obstacles are used as new sub-target nodes, and the global path is dynamically updated based on all the new sub-target points, generating a dynamic path that is more in line with the actual situation. This method effectively avoids the situation where the robot collides with obstacles or frequently stops and waits, and improves the overall operational efficiency of the warehouse.
[0056] Optionally, the step of reading all turning points on the global path of the storage and retrieval planning of the material box robot includes:
[0057] Extract all adjacent path nodes on the global path;
[0058] Direction analysis is performed on each of the adjacent path nodes to read all turning points on the global path of the material box robot's in-and-out storage planning.
[0059] The above extraction of all turning points on the global path and direction analysis thereof has the following technical benefits:
[0060] (1) By extracting all turning points on the global path, the system can more accurately understand where the robot needs to change direction during driving. This helps to more accurately consider the impact of turning points on the path in subsequent dynamic path planning, thereby generating a path that is more in line with the actual situation.
[0061] (2) Directional analysis of turning points can identify unnecessary turns or redundant driving paths. By optimizing these turning points, the system can generate a more concise and efficient driving path, reducing the robot's driving time and energy consumption.
[0062] (3) In complex warehouse environments, accurate extraction of turning points and direction analysis can help robots better adapt to changes in the environment. No matter how complex the warehouse layout is, the system can accurately identify turning points and generate adaptive dynamic paths based on them.
[0063] (4) Accurately handling turning points can reduce the robot’s sharp turns or sudden changes of direction during driving, thereby reducing the risk of collision with obstacles. This helps improve the robot’s driving safety in warehouse environments and protects the safety of the robot and goods.
[0064] (5) After extracting and analyzing the turning points, the system can more easily dynamically update the path according to the real-time situation during the subsequent driving process. When encountering moving obstacles or needing to change the driving route, the system can quickly adjust the turning points and generate a new dynamic path.
[0065] In addition, the following is a sample code to illustrate how to extract all turning points on the global path and perform direction analysis:
[0066]
[0067] Optionally, extracting all adjacent path nodes on the global path includes:
[0068] A coordinate point trajectory on a global path, based on which all adjacent path nodes on the global path are determined and extracted.
[0069] Optionally, the performing of direction analysis on each of the adjacent path nodes to read all turning points on the global path of the storage-in and storage-out planning of the material box robot includes:
[0070] Performing direction analysis on each of the adjacent path nodes to obtain a direction vector;
[0071] Based on the tolerance of the turning radius setting of the material box robot, the dot product and the modulus between two adjacent direction vectors are calculated to calculate the angle between the two adjacent direction vectors;
[0072] According to the angle, all turning points on the global path of the material box robot's storage and retrieval planning are determined and read.
[0073] To this end, the above solution extracts all adjacent path nodes on the global path and performs direction analysis on each adjacent path node, thereby achieving accurate reading of all turning points on the global path planned for the inbound and outbound planning of the material box robot, which has the following technical benefits:
[0074] (1) By performing directional analysis on adjacent path nodes on the global path, the turning points in the path can be accurately identified, thereby helping the container robot to more accurately plan the driving path for entering and exiting the warehouse, avoiding problems such as collisions or low driving efficiency caused by inaccurate path planning.
[0075] (2) Accurate reading of turning points is an important basis for autonomous navigation of robots. This technical solution determines the turning points by calculating the angle between adjacent direction vectors, enabling the bin robot to navigate autonomously and accurately in a complex warehouse environment, thus improving the robot's intelligence level.
[0076] (3) Accurately identifying turning points can help the box robot reduce unnecessary turns and stops during driving, thereby optimizing the driving path and improving driving efficiency. This is especially important for box robots that need to efficiently complete storage and retrieval tasks.
[0077] (4) The implementation of this technical solution makes the inbound and outbound planning of the material box robot more intelligent and automated, reduces the need for manual intervention, and improves the overall intelligent level of warehouse management.
[0078] Additionally, here is a sample code provided to illustrate how to extract adjacent path nodes on a global path and do a directional analysis to read turning points:
[0079]
[0080]
[0081] Optionally, extracting a plurality of key turning points from all the turning points, wherein one of the key turning points is used as a sub-target point, so as to form a sub-target point sequence according to the order of the key turning points on the corresponding global path, comprises:
[0082] The key turning point screening description is read to extract a number of key turning points from all the turning points based on the key turning point screening description, wherein one key turning point is used as a sub-target point to form a sub-target point sequence according to the sequence of the key turning points on the corresponding global path.
[0083] Optionally, the reading of the key turning point screening description to extract a plurality of key turning points from all the turning points based on the key turning point screening description, wherein one of the key turning points is used as a sub-target point to form a sub-target point sequence according to the sequence of the key turning points on the corresponding global path, includes:
[0084] Based on the set turning angle threshold, the distance threshold of adjacent turning points, and the path curvature threshold of the turn, several turning points whose turning angle is greater than the turning angle threshold, the distance between adjacent turning points is greater than the distance threshold, and the path curvature of the turn is greater than the path curvature threshold are extracted from all the turning points as several key turning points, one of the key turning points is used as a sub-target point to form a sub-target point sequence according to the sequence of the key turning points on the corresponding global path, one of the key turning points is used as a sub-target point to form a sub-target point sequence according to the sequence of the key turning points on the corresponding global path.
[0085] To this end, the above technical solution extracts several key turning points from all turning points, and forms a sub-target point sequence based on the sequence of these key turning points on the global path, thereby providing a more sophisticated and feasible path planning strategy for the inbound and outbound planning of the material box robot, which has the following technical benefits:
[0086] (1) By extracting key turning points and forming a sequence of sub-target points, the path planning problem can be simplified so that the robot only needs to consider these key nodes when planning the path, thereby greatly improving the efficiency of path planning.
[0087] (2) The extraction of key turning points is based on the set turning angle threshold, the distance threshold between adjacent turning points, and the turning path curvature threshold. This ensures that the extracted key turning points are feasible and important in the actual driving process of the robot, thereby enhancing the feasibility of path planning.
[0088] (3) By focusing on key turning points, the robot can handle these important path nodes more stably during driving, avoiding driving instability or path deviation caused by processing too many details.
[0089] (4) In practical applications, it may be necessary to adjust and optimize the path planning according to the actual driving conditions of the robot or changes in the warehouse environment. This technical solution facilitates such adjustment and optimization by extracting key turning points and forming a sequence of sub-target points.
[0090] In addition, the following is a sample code to illustrate how to extract key turning points from all turning points and form a sub-target point sequence:
[0091]
[0092]
[0093] Optionally, the forming of the sub-target point sequence according to the sequence of key turning points on the corresponding global path, wherein one key turning point is used as a sub-target point, and the forming of the sub-target point sequence according to the sequence of key turning points on the corresponding global path includes:
[0094] According to the sequence of the key turning points on the corresponding global path, each of the key turning points is numbered to obtain a path node;
[0095] Based on the size of the serial number, the path nodes are sorted, and the key turning points are added to the initialized target point sequence in sequence according to the sequence of the key turning points on the corresponding global path to form the sub-target point sequence.
[0096] To this end, the above technical solution provides a path planning method with a clear structure and easy management for the in and out planning of the material box robot by numbering and sorting the key turning points according to the order of their positions on the global path, and finally forming a sequence of sub-target points, which has the following technical benefits:
[0097] (1) By numbering and sorting key turning points, the path planning results are more intuitive and easier to understand. This not only facilitates debugging and testing for developers, but also makes subsequent path optimization and adjustment more convenient.
[0098] (2) The key turning points are formed into a sequence of sub-target points, and the nodes in the sequence can be flexibly adjusted according to actual needs, such as adding, deleting or modifying a sub-target point, without the need to reconstruct the entire path planning.
[0099] (3) This technical solution provides a universal path node processing method that can be easily integrated into the existing path planning system to improve the overall performance and maintainability of the system.
[0100] (4) Improving the robot’s driving efficiency: By optimizing the sub-target point sequence, the robot can process path nodes more efficiently during driving, reduce unnecessary driving and turning, and thus improve driving efficiency.
[0101] In addition, the following is a sample code to illustrate how to form a sub-target point sequence according to the order of key turning points on the global path:
[0102] #List of key turning points key_turning_points
[0103] key_turning_points=[(1,1),(3,1),(5,3)]
[0104] #Number the key turning points according to their sequence on the global path
[0105] numbered_key_points={}
[0106] for index,point in enumerate(key_turning_points):
[0107] numbered_key_points[point]=index
[0108] #Sort the path nodes based on the size of the numbers and form a sequence of sub-target points
[0109] sub_goal_sequence=sorted(numbered_key_points.items(),key=lambda x:x[1])
[0110] sub_goal_sequence=[point for point,index in sub_goal_sequence]
[0111] # Output sub-target point sequence
[0112] print("Sub-goal sequence:",sub_goal_sequence)
[0113] Optionally, performing sliding sampling on the sub-target point sequence based on a set sliding window to obtain a sampled target point group includes:
[0114] A speed sampling space constructed according to the moving speed limit and dynamic constraints of the material box robot is obtained, so as to perform sliding sampling on the sub-target point sequence based on a set sliding window to obtain a sampling target point group.
[0115] Optionally, the acquiring of a speed sampling space constructed according to a moving speed limit and a dynamic constraint of the material box robot, so as to obtain a sampling target point group by sliding sampling the sub-target point sequence based on a set sliding window, comprises:
[0116] A speed sampling space is obtained based on the moving speed limit of the material box robot and the dynamic constraint, wherein the moving speed limit includes the maximum and minimum values of the speed and acceleration, and the dynamic constraint is the maximum braking distance and the minimum braking distance when the material box robot encounters an obstacle.
[0117] According to the speed sampling space, the speed combination of the material box robot is predicted, so as to obtain a sampling target point group by sliding sampling of the sub-target point sequence based on a set sliding window.
[0118] To this end, the above technical solution obtains the speed sampling space constructed according to the moving speed limit and dynamic constraints of the material box robot, and performs sliding sampling on the sub-target point sequence based on the set sliding window to obtain the sampling target point group, which provides a more sophisticated and dynamic path planning strategy for the inbound and outbound planning of the material box robot, and has the following technical benefits:
[0119] (1) By considering the moving speed limit and dynamic constraints of the container robot, such as the maximum and minimum values of speed and acceleration, as well as the maximum and minimum braking distances when encountering obstacles, path planning can be more dynamically adapted to the actual driving conditions of the robot, thus improving the accuracy and feasibility of planning.
[0120] (2) By predicting the speed combination of the bin robot and considering dynamic constraints such as braking distance during the sliding sampling process, it can be ensured that the robot can safely respond to emergencies during driving, such as braking in time to avoid collision when encountering obstacles.
[0121] (3) By sampling the sub-target point sequence in a sliding window manner, a more streamlined and representative set of sampled target points can be obtained, which can not only reduce the amount of calculation for path planning, but also improve the efficiency of planning.
[0122] (4) The acquisition of the sampling target point group is based on the speed sampling space and the sliding window, which means that the path planning result can be optimized by adjusting the size of the speed sampling space or the sliding window, making the planning more flexible and controllable.
[0123] In addition, the following is a sample code to illustrate how to perform sliding sampling on the sub-target point sequence based on the set sliding window to obtain the sampled target point group:
[0124]
[0125] Optionally, predicting the speed combination of the material box robot according to the speed sampling space, so as to obtain a sampling target point group by sliding sampling the sub-target point sequence based on a set sliding window, comprises:
[0126] According to the speed sampling space, the speed combination of the material box robot is predicted, and the speed combination is mapped to the set sliding window to determine from the sub-target point sequence that the material box robot can reach the sub-target point or the vicinity of the sub-target point within the time of the sliding window, and the sub-target point that is reached or the sub-target point that satisfies the vicinity is used as the sampling target point to form a sampling target point group.
[0127] To this end, the speed combinations of the material box robot are predicted according to the speed sampling space, and these speed combinations are mapped to the set sliding window, so as to determine the sub-target points or the vicinity of the sub-target points that the material box robot can reach within the time of the sliding window from the sub-target point sequence, thereby forming a sampling target point group, which has the following technical benefits:
[0128] (1) By predicting the speed combinations of the container robot and considering the actual driving effects of these speed combinations within the sliding window time, the sub-target points that the robot can reach in the future can be determined more accurately, thereby improving the foresight and accuracy of path planning.
[0129] (2) The speed sampling space contains a variety of speed and acceleration combinations of the container robot, which enables path planning to more dynamically adapt to different driving scenarios and constraints. For example, when encountering obstacles or needing to accelerate through a certain path, the appropriate speed combination can be selected to achieve this.
[0130] (3) Sampling the sub-target point sequence by means of a sliding window can greatly reduce the amount of calculation for path planning. At the same time, since the sampled target point group is obtained based on the velocity sampling space and the sliding window prediction, the sub-target points that have an important impact on the path planning results can be selected more specifically, further improving the planning efficiency.
[0131] (4) The acquisition of the sampling target point group is based on the prediction results of the speed sampling space and the sliding window, which means that the path planning result can be optimized by adjusting the size of the speed sampling space or the sliding window. At the same time, since the sampling target point group is representative, the entire path planning result can also be fine-tuned by adjusting these points.
[0132] In addition, the following is a sample code to illustrate how to predict the speed combination of the container robot based on the speed sampling space, and perform sliding sampling on the sub-target point sequence based on the set sliding window to obtain the sampling target point group:
[0133]
[0134]
[0135] Optionally, the determining whether a moving obstacle appears near the sampling target point group, and if so, adjusting the dynamic description feature of the material box robot, and using the moving obstacle as a newly added sub-target node, includes:
[0136] The material box robot is subjected to environmental perception when moving along a static path to determine dynamic obstacles with a set distance similarity to the sampling target point group. If such obstacles appear, it is determined that a moving obstacle appears near the sampling target point group. If such obstacles appear, at least one of the moving speed and the moving direction of the material box robot is adjusted, and the moving obstacle is added as a new sub-target node. The dynamic description features of the material box robot include at least one of the moving speed and the moving direction.
[0137] To this end, by judging whether there are moving obstacles near the sampling target point group, and adjusting the dynamic description features of the box robot when they appear, and taking the moving obstacles as new sub-target nodes, it has the following technical benefits:
[0138] (1) Through real-time environmental perception, moving obstacles near the sampling target point group can be discovered in time, thereby preventing the box robot from colliding with these obstacles during its movement, greatly improving the safety of path planning.
[0139] (2) When a moving obstacle is detected, the dynamic description characteristics of the material box robot, such as its moving speed and direction, can be adjusted in real time, allowing the robot to flexibly respond to environmental changes and enhancing the flexibility and adaptability of path planning.
[0140] (3) By treating moving obstacles as newly added sub-target nodes, the dynamic changes of obstacles can be considered in path planning, thereby avoiding frequent replanning of the entire path and improving the efficiency of path planning.
[0141] (4) By adjusting the dynamic description features of the container robot in real time and adding moving obstacles as new sub-target nodes, the path planning results can be optimized and adjusted more finely, making the planning results more consistent with actual driving conditions.
[0142] In addition, the following is a sample code that illustrates how to determine whether there are moving obstacles near the sampling target point group, and adjust the dynamic description features of the container robot when they appear, while adding the moving obstacles as new sub-target nodes:
[0143]
[0144]
[0145] Optionally, the adjusting at least one of the moving speed and the moving direction of the material box robot and taking the moving obstacle as a newly added sub-target node includes:
[0146] Based on the set evaluation function, the current dynamic characteristics of the material box robot are evaluated to obtain a dynamic evaluation value, so as to adjust at least one of the moving speed and the travel direction of the material box robot according to the dynamic evaluation value, and the moving obstacle is used as a new sub-target node. The evaluation function includes: a direction angle evaluation sub-function, which represents the direction angle deviation of the moving obstacle relative to the current target point at the current speed; a distance evaluation sub-function, which represents the distance deviation of the moving obstacle relative to the current target point at the current speed; and a current speed magnitude evaluation sub-function, which represents the magnitude of the current speed.
[0147] To this end, this technical solution evaluates the current dynamic characteristics of the material box robot based on the set evaluation function, and adjusts the robot's moving speed and direction according to the obtained dynamic evaluation value. At the same time, the moving obstacles are used as new sub-target nodes, which has the following technical benefits:
[0148] (1) By quantitatively evaluating the current dynamic characteristics of the material box robot through the evaluation function, it can more intelligently and autonomously determine whether the robot needs to adjust its moving speed or direction, and whether it needs to add moving obstacles as new sub-target nodes, thereby improving the intelligence and autonomy of path planning.
[0149] (2) The evaluation function takes into account multiple factors such as angular deviation, distance deviation, and current speed, and can reflect the relative relationship between the container robot and the moving obstacle in real time, thereby adjusting the dynamic description characteristics of the robot according to the real-time situation, thereby enhancing the real-time and adaptability of path planning.
[0150] (3) By adjusting the moving speed and direction of the material box robot and adding moving obstacles as new sub-target nodes, the robot can be more effectively prevented from colliding with obstacles, thereby improving the safety of path planning. At the same time, since the evaluation function can reflect the relative relationship between the robot and the obstacle in real time, the dynamic description characteristics of the robot can be adjusted more specifically, thereby improving the efficiency of path planning.
[0151] (4) Each sub-function in the evaluation function can be customized and optimized according to the actual application scenario to meet different path planning requirements. For example, in some scenarios, the direction angle deviation may be more concerned, while in other scenarios, the distance deviation or the current speed may be more concerned. This flexibility allows path planning to be more personalized to meet actual application needs.
[0152] In addition, the following is a sample code that illustrates how to evaluate the current dynamic characteristics of the container robot based on the set evaluation function, and adjust the robot's moving speed and direction according to the obtained dynamic evaluation value, while taking the moving obstacle as a new sub-target node:
[0153]
[0154]
[0155] Optionally, dynamically updating the global path based on all the newly added sub-target points determined to generate a dynamic path for the container robot to enter and exit the warehouse includes:
[0156] According to the degree of proximity between each newly added sub-target point and the current target point position of the material box robot, the free running position points that the material box robot can reach to avoid the newly added sub-target points are determined from the free running area, so that the global path is dynamically updated based on the free running position points to generate a dynamic path for the material box robot to enter and exit the warehouse.
[0157] To this end, this technical solution dynamically updates the global path based on all the newly added sub-target points to generate a dynamic path for the bin robot to enter and exit the warehouse, which has the following technical benefits:
[0158] (1) By dynamically updating the global path according to the newly added sub-target points, the container robot can adapt to environmental changes more flexibly, adjust the in-and-out paths in a timely manner, and avoid collisions with obstacles.
[0159] (2) When dynamically updating the global path, the proximity between the newly added sub-target point and the current target point of the material box robot is taken into account, and the free running position point that the material box robot can reach in order to avoid the newly added sub-target point is determined from the free running area. In this way, the material box robot can always maintain a safe distance during the process of entering and leaving the warehouse, avoiding collisions with obstacles or newly added sub-target points, thereby improving the safety and reliability of path planning.
[0160] (3) By dynamically updating the global path, the container robot can complete the in-and-out tasks more efficiently. When encountering a new sub-target point, the robot can quickly adjust the path to avoid unnecessary detours or stagnation, thereby improving the efficiency of in-and-out.
[0161] (4) This technical solution uses an algorithm to dynamically update the path without human intervention, making it easier to automate and intelligentize path planning. This helps reduce labor costs and improve the level of automation in warehousing logistics.
[0162] In addition, the following is a sample code that illustrates how to dynamically update the global path based on all the newly added sub-target points to generate a dynamic path for the bin robot to enter and exit the warehouse:
[0163]
[0164]
[0165] This code first sets the global path, the newly added sub-target points, and the free running area. Then, it defines an update_global_path function to dynamically update the global path based on the newly added sub-target points. This function traverses each point in the global path and checks whether there are newly added sub-target points between the current point and the next point. If there are, it finds the free running position point that avoids the newly added sub-target points and updates the path. Finally, it prints out the updated global path.
[0166] Optionally, the step of reading all turning points on the global path of the storage and retrieval planning of the material box robot includes:
[0167] Reading a global environment model of the warehouse where the material box robot is located, wherein the global environment model defines an obstacle area and a free running area;
[0168] Parsing the in-and-out instructions of the material box robot to determine the current starting position and the current target point position of the material box robot;
[0169] According to the starting point position, the current target point position and the global environment model, a global path search is performed for the material box robot to enter and exit the warehouse, and a plurality of optional global paths are obtained to form an optional global path set;
[0170] Based on the set global path constraints, the optional global path set is solved to select an optional global path that meets the set path constraints and uses it as the planned global path.
[0171] The above-planned global path has the following technical benefits:
[0172] (1) By reading the global environment model of the warehouse where the container robot is located, the system can accurately understand the obstacle area and the free running area, providing reliable environmental information for subsequent path planning. This helps to avoid collisions between the robot and obstacles during driving and improve driving safety.
[0173] (2) By parsing the in-and-out instructions of the material box robot, the robot’s current starting position and target point position can be accurately determined. This ensures the accuracy and effectiveness of path planning, allowing the robot to accurately reach the target point along the predetermined path.
[0174] (3) The system can search for multiple optional global paths based on the starting point location, the target point location, and the global environment model, and form a set of optional global paths. This provides diversity in path selection and helps find the optimal or suboptimal path solution.
[0175] (4) Based on the set global path constraints, the system can select a path that satisfies the constraints from the optional global path set as the planned global path. This ensures the feasibility and optimality of the selected path, enabling the robot to complete the task efficiently while satisfying the constraints.
[0176] (5) By extracting key turning points and forming a sequence of sub-target points, the system can finely process turning points on the global path, reducing unnecessary turns and driving distance. At the same time, the system can quickly respond to moving obstacles, dynamically update the global path, and generate a dynamic path that is more in line with the actual situation.
[0177] In addition, the following is an example code to illustrate how to read the global environment model and perform path search:
[0178]
[0179]
[0180] Optionally, the inbound and outbound instructions include inbound instructions and / or outbound instructions, so as to complete an outbound or inbound task according to the task order, and multiple goods may be taken or unloaded at one time;
[0181] The warehousing instruction is used to schedule the material box robot to place the goods from the starting position to the warehousing workbench; correspondingly, the optional static path is: the route of the material box robot from its optional starting point position to the warehousing workbench, from the warehousing workbench to the first shelf, and from the first shelf to the next shelf until all the goods are placed;
[0182] The warehousing instruction is used to schedule the material box robot to pick up goods from the starting position to the warehousing workbench; correspondingly, the optional static path includes: the route of the material box robot from the starting position to the first shelf, the route from the first shelf to the next shelf until the goods required for the order are picked up, and the route from the last shelf to the outbound workbench.
[0183] Optionally, the reading of a global environment model of a warehouse where the material box robot is located, wherein the global environment model defines an obstacle area and a free running area, further includes:
[0184] Obtaining a global environment model file of the warehouse where the material box robot is located, parsing the global environment model file, and determining the central coordinates of obstacles in the warehouse and the outlines of the obstacles;
[0185] A global environment model of the warehouse is constructed based on the central coordinates of the obstacle and the outline of the obstacle.
[0186] Optionally, constructing the global environment model of the warehouse based on the central coordinates of the obstacle and the outline of the obstacle includes:
[0187] Gridding the area occupied by the warehouse to obtain a plurality of grids to form a grid array;
[0188] Based on the central coordinates of the obstacle in the warehouse, matching the outline of the obstacle with the grid array to determine a mapping relationship between the obstacle and the grid array;
[0189] Based on the mapping relationship, the grid array is encoded to obtain obstacle position encoding and free area position encoding;
[0190] The obstacle position codes and free area position codes are projected onto the grid array to form a global environment model of the warehouse.
[0191] Optionally, constructing the global environment model of the warehouse based on the central coordinates of the obstacle and the outline of the obstacle includes:
[0192] Based on the center coordinates of the obstacle and the outline of the obstacle, defining a plurality of designated position coordinates that the material box robot will pass through when traveling in the warehouse to form a designated position coordinate sequence;
[0193] Determine the route of the material box robot from its starting position point coordinates to the nearest specified position coordinates, and the route from the nearest specified position coordinates to the next specified position coordinates, iterating in sequence to obtain a plurality of routes;
[0194] Map each route as an edge and mark obstacles on each edge;
[0195] All edges loaded with obstacle identifiers are spliced together in the order of iteration to form a global environment model of the warehouse.
[0196] Optionally, constructing the global environment model of the warehouse based on the central coordinates of the obstacle and the outline of the obstacle includes:
[0197] Generating polygons representing different obstacles based on the central coordinates of the obstacles and the outlines of the obstacles;
[0198] Determine the coordinates of each vertex of each polygon to establish a line connecting the optional starting point position and all vertices, to establish a line connecting the optional target point position and all vertices, and a line connecting vertices of different polygons;
[0199] The optional starting point position, the optional target point position, the polygon, the line connecting the optional starting point position and all vertices, the line connecting the optional target point position and all vertices, and the line connecting the vertices of different polygons are projected onto the area occupied by the warehouse to form a global environment model of the warehouse.
[0200] Optionally, the parsing of the storage in and out instruction of the material box robot to determine the current starting position and the current target point position of the material box robot includes:
[0201] Parsing the in-and-out instructions of the material box robot to obtain in-and-out configuration parameters;
[0202] According to the outbound parameters, it is determined whether it is an outbound instruction or an inbound instruction, and the current starting point position and the current target point position of the corresponding material box robot.
[0203] Optionally, the static path search for the in-and-out storage of the material box robot is performed according to the current starting point position, the current target point position and the global environment model to obtain a plurality of optional static paths to form an optional static path set, including:
[0204] According to the starting point position, the current target point position and the global environment model, a number of particles are constructed, and an initial fitness value and an initial inertia weight are assigned to each particle to obtain an initialized particle group, each particle representing an initialized static path;
[0205] The initialized particle swarm is iterated to search for a static path for the material box robot to enter and exit the warehouse, and a plurality of optional static paths are obtained to form an optional static path set.
[0206] Optionally, the initial fitness value represents the initial performance strength of the corresponding initialized static path, and the initial performance strength includes the length of the path, the initial number of turns made by the material box robot when moving along the initialized static path, and the average safety distance from obstacles.
[0207] Optionally, the iterating of the initialized particle swarm to search for a static path for the material box robot to enter and exit the warehouse, and obtain a number of optional static paths to form an optional static path set, including: based on a set path search mechanism, determining a self-cognition metric and a social cognition metric to be used when iterating the initialized particle swarm, the self-cognition metric is used to represent the initialized semantic distance between the current path length of the particle and its shortest path length, and the social cognition metric is used to represent the initialized semantic distance between the current path length of the particle and the global shortest path length; based on the self-cognition metric and the social cognition metric, iterating the initialized particle swarm to search for a static path for the material box robot to enter and exit the warehouse, and obtain a number of optional static paths to form an optional static path set.
[0208] Optionally, the iterating of the initialization particle swarm based on the self-cognition metric and the social cognition metric to search for a static path for the material box robot to enter and exit the warehouse, and obtain a number of optional static paths to form an optional static path set, including: based on the self-cognition metric and the social cognition metric, in the space where the initialization particle swarm is located, iterating the initialization particle swarm to update the particles in the initialization particle swarm in real time to obtain dynamic particles, so as to obtain a dynamic particle swarm, each dynamic particle having an updated fitness value and an updated inertia weight; based on the updated fitness value, filtering out some dynamic particles from the dynamic particle swarm to search for a static path for the material box robot to enter and exit the warehouse, and obtain a number of optional static paths to form an optional static path set.
[0209] Optionally, the method further includes: linearly updating the inertia weights of the particles in the initialized particle swarm according to the inertia weight update rule, so that the inertia weights decrease linearly with the increase of the number of iterations, and the linear decrease makes it possible to search for static paths for entering and exiting the warehouse based on the nearest search position and search direction of the particles in the space where the initialized particle swarm is located.
[0210] Optionally, based on the set static path constraint, the optional static path set is solved to select an optional static path that satisfies the set path constraint as the planned static path, including: the static path constraint is set at a fitness threshold, and the updated fitness value of the updated particle corresponding to each static optional path in the optional static path set is globally solved to select an optional static path whose updated path length, number of turns, and average safety distance from obstacles globally match the fitness threshold as the planned static path.
[0211] The above description is only a preferred embodiment of the present application and an explanation of the technical principles used. Those skilled in the art should understand that the scope of the invention involved in the present application is not limited to the technical solution formed by a specific combination of the above technical features, but should also cover other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the above invention concept. For example, the above features are replaced with the technical features with similar functions disclosed in this application (but not limited to) by each other to form a technical solution.
Claims
1. A dynamic path planning method for a material box robot entering and exiting the warehouse, It is characterized in that include: Read all turning points on the global path planned by the container robot for inbound and outbound storage; Extracting a number of key turning points from all the turning points, wherein one of the key turning points is used as a sub-target point, so as to form a sub-target point sequence according to the order of the key turning points on the corresponding global path; Perform sliding sampling on the sub-target point sequence based on a set sliding window to obtain a sampling target point group; Determine whether a moving obstacle appears near the sampling target point group. If so, adjust the dynamic description feature of the material box robot and use the moving obstacle as a newly added sub-target node; Based on all the newly added sub-target points determined, the global path is dynamically updated to generate a dynamic path for the container robot to enter and exit the warehouse.
2. A dynamic path planning method for a material box robot entering and exiting a warehouse according to claim 1, It is characterized in that The reading of all turning points on the global path of the material box robot in and out of the warehouse planning includes: Extract all adjacent path nodes on the global path; Performing direction analysis on each of the adjacent path nodes to read all turning points on the global path planned for the inbound and outbound storage of the material box robot; The step of extracting all adjacent path nodes on the global path includes: A coordinate point trajectory on a global path, based on which all adjacent path nodes on the global path are determined and extracted; The direction analysis of each adjacent path node to read all turning points on the global path of the storage and warehousing planning of the container robot includes: Performing direction analysis on each of the adjacent path nodes to obtain a direction vector; Based on the tolerance of the turning radius setting of the material box robot, the dot product and the modulus between two adjacent direction vectors are calculated to calculate the angle between the two adjacent direction vectors; According to the angle, all turning points on the global path of the material box robot's storage and retrieval planning are determined and read.
3. A dynamic path planning method for a material box robot entering and exiting a warehouse according to claim 1, It is characterized in that The step of extracting a plurality of key turning points from all the turning points, wherein one of the key turning points is used as a sub-target point, so as to form a sub-target point sequence according to the sequence of the key turning points on the corresponding global path, includes: The key turning point screening description is read to extract a number of key turning points from all the turning points based on the key turning point screening description, wherein one key turning point is used as a sub-target point to form a sub-target point sequence according to the sequence of the key turning points on the corresponding global path.
4. A dynamic path planning method for a material box robot entering and exiting a warehouse according to claim 3, It is characterized in that The reading of the key turning point screening description to extract a plurality of key turning points from all the turning points based on the key turning point screening description, wherein one of the key turning points is used as a sub-target point to form a sub-target point sequence according to the sequence of the key turning points on the corresponding global path, includes: Based on the set turning angle threshold, the distance threshold of adjacent turning points, and the path curvature threshold of the turn, several turning points whose turning angle is greater than the turning angle threshold, the distance between adjacent turning points is greater than the distance threshold, and the path curvature of the turn is greater than the path curvature threshold are extracted from all the turning points as several key turning points, one of the key turning points is used as a sub-target point to form a sub-target point sequence according to the sequence of the key turning points on the corresponding global path, one of the key turning points is used as a sub-target point to form a sub-target point sequence according to the sequence of the key turning points on the corresponding global path.
5. A dynamic path planning method for a material box robot entering and exiting a warehouse according to claim 3 or 4, It is characterized in that The forming of a sub-target point sequence according to the sequence of key turning points on the corresponding global path, wherein one key turning point is used as a sub-target point, and the forming of a sub-target point sequence according to the sequence of key turning points on the corresponding global path includes: According to the sequence of the key turning points on the corresponding global path, each of the key turning points is numbered to obtain a path node; Based on the size of the serial number, the path nodes are sorted, and the key turning points are added to the initialized target point sequence in sequence according to the sequence of the key turning points on the corresponding global path to form the sub-target point sequence.
6. A dynamic path planning method for a material box robot entering and exiting a warehouse according to claim 1, It is characterized in that The step of performing sliding sampling on the sub-target point sequence based on a set sliding window to obtain a sampling target point group includes: A speed sampling space constructed according to the moving speed limit and dynamic constraints of the material box robot is obtained, so as to perform sliding sampling on the sub-target point sequence based on a set sliding window to obtain a sampling target point group.
7. A dynamic path planning method for a material box robot entering and exiting a warehouse according to claim 6, It is characterized in that The acquiring of the speed sampling space constructed according to the moving speed limit and dynamic constraints of the material box robot, and performing sliding sampling on the sub-target point sequence based on a set sliding window to obtain a sampling target point group, includes: A speed sampling space is obtained based on the moving speed limit of the material box robot and the dynamic constraint, wherein the moving speed limit includes the maximum and minimum values of the speed and acceleration, and the dynamic constraint is the maximum braking distance and the minimum braking distance when the material box robot encounters an obstacle. According to the speed sampling space, the speed combination of the material box robot is predicted, so as to obtain a sampling target point group by sliding sampling of the sub-target point sequence based on a set sliding window.
8. A dynamic path planning method for a material box robot entering and exiting a warehouse according to claim 10, It is characterized in that The predicting of the speed combination of the material box robot according to the speed sampling space, and performing sliding sampling on the sub-target point sequence based on a set sliding window to obtain a sampling target point group, includes: According to the speed sampling space, the speed combination of the material box robot is predicted, and the speed combination is mapped to the set sliding window to determine from the sub-target point sequence that the material box robot can reach the sub-target point or the vicinity of the sub-target point within the time of the sliding window, and the sub-target point that is reached or the sub-target point that satisfies the vicinity is used as the sampling target point to form a sampling target point group.
9. A dynamic path planning method for a material box robot entering and exiting a warehouse according to claim 1, It is characterized in that The determining whether a moving obstacle appears near the sampling target point group, and if so, adjusting the dynamic description feature of the material box robot, and using the moving obstacle as a newly added sub-target node, includes: The material box robot is subjected to environmental perception when moving along a static path to determine dynamic obstacles with a set distance similarity to the sampling target point group. If such obstacles appear, it is determined that a moving obstacle appears near the sampling target point group. If such obstacles appear, at least one of the moving speed and the moving direction of the material box robot is adjusted, and the moving obstacle is added as a new sub-target node. The dynamic description features of the material box robot include at least one of the moving speed and the moving direction.
10. A dynamic path planning method for a material box robot entering and exiting a warehouse according to claim 1, It is characterized in that The step of adjusting at least one of the moving speed and the moving direction of the material box robot and using the moving obstacle as a newly added sub-target node includes: Based on the set evaluation function, the current dynamic characteristics of the material box robot are evaluated to obtain a dynamic evaluation value, so as to adjust at least one of the moving speed and the travel direction of the material box robot according to the dynamic evaluation value, and the moving obstacle is used as a new sub-target node. The evaluation function includes: a direction angle evaluation sub-function, which represents the direction angle deviation of the moving obstacle relative to the current target point at the current speed; a distance evaluation sub-function, which represents the distance deviation of the moving obstacle relative to the current target point at the current speed; and a current speed magnitude evaluation sub-function, which represents the magnitude of the current speed.
Citation Information
Patent Citations
Robot dynamic obstacle avoidance method based on artificial potential field and rolling window
CN108762264A
Robot trajectory planning optimal control method based on obstacle size homotopy strategy
CN111267110A
Rapid ant colony unmanned ship dynamic obstacle avoidance method based on danger index and distance function
CN114625150A
Mobile robot path planning optimization method
CN115167474A
CTU path planning method fusing particle swarm optimization and dynamic window method
CN116203950A
