Intelligent storage robot path planning method based on TEB dynamic obstacle avoidance and improved bidirectional A-Star algorithm

Abstract

The invention discloses an intelligent storage robot path planning method, and belongs to the technical field of intelligent robots, and the method comprises the steps: improving a bidirectional A-Star algorithm, and combining a Time Elasticic-Band (TEB) obstacle avoidance algorithm, and providing a new intelligent storage path planning method. The improved A-Star algorithm adopts bidirectional search to replace unidirectional search of a traditional A-Star algorithm; 5-neighborhood search is adopted to replace 8-neighborhood search; introducing a self-adaptive dynamic search factor to improve a heuristic function; and smoothing the path by adopting a redundant node removal and linear interpolation method. The improved bidirectional A-Star algorithm searches to obtain a global optimal path, and the TEB algorithm optimizes and adjusts a local path according to the global path to complete dynamic obstacle avoidance. According to the method, the speed and precision of path planning are effectively improved, the local path can be adjusted in real time to avoid dynamic obstacles, and the requirement for path planning of the intelligent storage robot is met.

Description

Technical Field

[0001] The present invention discloses a path planning method for an intelligent warehousing robot, belonging to the technical field of intelligent robots. Background Technique

[0002] With the wide application of intelligent warehousing systems, path planning, as one of the important technologies for realizing automated warehousing management, has received great attention. Path planning is one of the basic problems in robotics and is particularly important in the autonomous navigation system of mobile robots. In an intelligent warehousing environment, the goal of path planning is to find a feasible and optimal path between the starting point and the target point through scientific algorithms, while avoiding obstacles during the driving process. Path planning is divided into two major categories: global path planning and local path planning. Global path planning focuses on planning an overall optimal path for the entire warehousing system. Common algorithms include the Dijkstra algorithm, the A-Star algorithm, the rapidly-exploring random tree algorithm, etc. Local path planning pays more attention to real-time decision-making and adjustment in a local area, especially in dealing with obstacles or unpredictable changes in a dynamic environment. Common algorithms include the dynamic window method, the artificial potential field method, the time elastic band method, etc.

[0003] As a classic graph search algorithm, the A-Star algorithm is widely used in path planning problems due to its high efficiency and accuracy. This algorithm combines the actual cost from the starting point to the current point with the estimated cost from the current point to the target point through a heuristic search method, and can quickly find the optimal path in a static environment. In intelligent warehousing, the A-Star algorithm is commonly used for global path planning and is particularly suitable for environments where obstacles are known and static. However, the A-Star algorithm has obvious deficiencies in dealing with local obstacle avoidance problems in a dynamic environment. As the obstacles change, the A-Star path needs to be recalculated and cannot avoid new obstacles in real time. Therefore, it cannot fully meet the requirements of a dynamic environment in intelligent warehousing.

[0004] In order to overcome the limitations of the A-Star algorithm, a local obstacle avoidance algorithm needs to be introduced. Among them, the TEB algorithm, as a commonly used local path optimization algorithm, has been widely applied. The TEB algorithm is based on the method of trajectory optimization, can adjust the robot path in real time to avoid dynamic obstacles, and takes into account kinematic constraints, speed and acceleration limits and other factors, thus ensuring the smoothness and safety of the path. In an intelligent warehousing system, the TEB algorithm can be used to achieve real-time obstacle avoidance, especially in complex environments and narrow channels, and can effectively avoid obstacles and optimize the motion trajectory. However, although the TEB algorithm performs excellently in local path adjustment, its ability in global path planning is weak. Therefore, when the environment changes greatly, the TEB cannot provide the ability to re-plan the path as a whole.

[0005] Although the A-Star algorithm and the TEB algorithm have certain advantages in their respective application scenarios, they also have obvious deficiencies. Although the A-Star algorithm can provide a globally optimal path, when facing changes in obstacles in a dynamic environment, it cannot make real-time adjustments, resulting in frequent recalculation of the path and low efficiency. The TEB algorithm, on the other hand, can effectively perform local obstacle avoidance and real-time path adjustment, but lacks the ability of global path planning, and its adjustment effect is limited when facing large environmental changes. Therefore, how to combine the advantages of these two algorithms to ensure the accuracy of global path planning and effectively perform local obstacle avoidance in a dynamic environment has become the key to improving the path planning performance of intelligent warehousing. Summary of the Invention

[0006] The main object of the present invention is: in view of the challenges of path planning in an intelligent warehousing environment, the present invention proposes an intelligent warehousing robot path planning method based on TEB dynamic obstacle avoidance and improved bidirectional A-Star algorithm, making full use of the advantages of the A-Star algorithm in planning the optimal global path and the efficient dynamic obstacle avoidance and local path adjustment capabilities of the TEB algorithm. This method can achieve real-time obstacle avoidance for dynamic obstacles and effectively cope with complex changes in the warehousing environment on the basis of ensuring the accuracy of global path planning.

[0007] In order to achieve the above object, an intelligent warehousing robot path planning method based on TEB dynamic obstacle avoidance and improved bidirectional A-Star algorithm studied by the present invention includes the following steps:

[0008] S1: The autonomous mobile robot builds a map according to the current environment and rasterizes it, obtains information such as obstacles according to the map content, and marks the obstacle area and the drivable area for subsequent path planning.

[0009] S2: Initialize relevant parameters. Such as defining the starting point and the ending point, and initializing the node list.

[0010] S3: Calculate the adaptive dynamic factor according to the relationship between multiple nodes, introduce it into the heuristic function, and dynamically adjust the weight of the heuristic function.

[0011] S4: Perform bidirectional search according to two target points, select the expansion node according to the angle data between the parent nodes of the two paths, and select the node with the smallest cost value for expansion until the optimal path is found.

[0012] S5: Remove redundant nodes from the planned optimal path, and then use the linear interpolation method to smooth the path.

[0013] S6: Provide the smoothed global path to the TEB algorithm for local path planning to achieve the effect of dynamic obstacle avoidance, and plan a path that meets the kinematic requirements and feasibility of the robot.

[0014] An intelligent warehousing robot path planning method based on TEB dynamic obstacle avoidance and improved bidirectional A-Star algorithm as described above, characterized in that the operation of S1 is specifically as follows: The autonomous mobile robot collects the surrounding environment information of the warehousing area through radar and dynamically builds a map using gmapping. The built map is rasterized in a two-dimensional plane, and each grid represents a unit area. The obstacles in the environment are detected and calibrated, with the obstacle area represented by "1" and the drivable area represented by "0".

[0015] An intelligent warehousing robot path planning method based on TEB dynamic obstacle avoidance and improved bidirectional A-Star algorithm as described above, characterized in that the operation of S2 is specifically as follows: Initialize the cost, map parameters, define the target point and the starting point, and initialize two Openlist open node lists and two Closelist closed node lists respectively. Put the target point and the starting point into the Openlist1 list and the Openlist2 list respectively, corresponding to the paths path1 and path2 respectively.

[0016] An intelligent warehousing robot path planning method based on TEB dynamic obstacle avoidance and improved bidirectional A-Star algorithm as described above, characterized in that the operation of S3 is specifically as follows: In the heuristic function, the actual cost of the robot's travel is represented by the Euclidean distance, and the estimated cost is represented by the Manhattan distance. Using these two distances to calculate the total cost together better meets the path requirements of the warehousing environment. Calculate the adaptive dynamic factor according to the relationship between multiple nodes and dynamically adjust the weight of the estimated cost. The specific calculation method is as follows:

[0017] (1) Select the Euclidean distance and the Manhattan distance to calculate the cost, and the formula is expressed as:

[0018] f(i) = g(i) + h(i) (1)

[0019]

[0020] h(i) = |x t -x i | + |y t -y i | (3)

[0021] Where i represents the currently searched node, g(i) represents the actual cost from the starting node to node i, h(i) is the estimated cost from node i to the target node, usually represented by the remaining distance, and f(i) represents the total cost of the current node. (x i , y i ) represents the coordinates of node i, (x s , y s) represents the starting point coordinates, (x t , y t ) represents the target point coordinates.

[0022] (2) Calculate the relevant dynamic factor coefficients based on the starting point, ending point, and the parent nodes of the two paths. The expression is:

[0023]

[0024] Define path1 as the path searched starting from the starting point, and path2 as the path searched starting from the target point. Among them, (x p1 , y p1 ) and (x p2 , y p2 ) are the search parent nodes of path1 and path2 respectively. ξ 1 and ξ 5 are the distances from the starting point to the two parent nodes respectively, ξ 3 and ξ 6 are the distances from the target point to the two parent nodes respectively, ξ 2 is the distance between the two parent nodes, and ξ 4 is the distance from the starting point to the target point. Then the improved heuristic functions of path1 and path2 are:

[0025] f 1 (i) = g 1 (i) + (0.2 * δ 1 + 0.8 * α 1 ) * h 1 (i) (5)

[0026] f 2 (i) = g 2 (i) + (0.2 * δ 2 + 0.8 * α 2 ) * h 2 (i) (6)

[0027] Among them, δ 1 is determined by the ratio of the sum of ξ 1 and ξ 2 to ξ 1 , and δ 2 is calculated from the ratio of the sum of ξ 2 and ξ 3 to ξ 3 . α 1 is determined by the ratio of ξ 4 and ξ 5 , and α 2 is determined by the ratio of ξ 4 and ξ 6 .

[0028] An intelligent warehouse robot path planning method based on TEB dynamic obstacle avoidance and improved bidirectional A-Star algorithm as described above, characterized in that the operation of S4 is specifically as follows: Define path1 to start searching from the starting point and path2 to start searching from the end point, and the two paths are searched simultaneously. During the search process, calculate the angle between the two paths based on the coordinates of the parent nodes of the two paths, and the parent nodes of the two paths select the nodes to be expanded according to the angle value. Select the node with the smallest cost value for expansion, and repeat this step. When the two parent nodes meet, the path planning task is completed.

[0029] An intelligent warehouse robot path planning method based on TEB dynamic obstacle avoidance and improved bidirectional A-Star algorithm as described above, characterized in that the operation of S5 is specifically as follows: Process the path generated in S4. The path planned in S4 usually contains redundant nodes. Therefore, remove the redundant nodes from this path to reduce the turning of the path and make it more concise. The path after removing the redundant nodes is not smooth enough and thus does not meet the kinematic constraints of the robot. Therefore, use the linear interpolation method to smooth the path.

[0030] An intelligent warehouse robot path planning method based on TEB dynamic obstacle avoidance and improved bidirectional A-Star algorithm as described above, characterized in that the operation of S6 is specifically as follows: Provide the optimized global path in S5 to the TEB algorithm. The TEB algorithm decomposes the global path into multiple local paths and optimizes the local paths according to constraints such as time optimality and kinematics. When encountering a dynamic obstacle during the robot's driving process and affecting its driving, adjust the local path according to the obstacle avoidance constraint conditions to avoid the dynamic obstacle and continue driving.

[0031] Compared with the prior art, the beneficial effects of the present invention are:

[0032] 1) Adopt bidirectional search and 5-neighborhood search, search from the starting point and the target point simultaneously, and select the expansion node according to the angle between the two target points. Compared with the traditional unidirectional search and 8-neighborhood search, it reduces the search space and computational complexity;

[0033] 2) Adopt redundant node removal and linear interpolation method to simplify and smooth the planned global path, so that the route meets the kinematic requirements of the robot;

[0034] 3) Introduce a dynamic adaptive weight factor, dynamically adjust the heuristic function according to the distance between relevant nodes, and can improve the path search speed and accuracy compared with the traditional heuristic function;

[0035] 4) Combine the TEB algorithm, combine global path planning with real-time local obstacle avoidance, the path is smoother and more efficient compared with the APF algorithm, and avoid the influence of sudden speed changes. Ensure the safety and efficient navigation of the robot in a dynamic environment.

[0036] To make the above objects, features, and advantages of the present invention more apparent and understandable, specific embodiments of the present invention are hereinafter given, and in conjunction with the accompanying drawings, the detailed description is as follows. Brief Description of the Drawings

[0037] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the embodiments will be briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present invention and should not be regarded as a limitation of the scope. For those of ordinary skill in the art, without creative efforts, other related drawings can also be obtained based on these drawings.

[0038] Figure 1 It is a flowchart of the path planning method in the present invention;

[0039] Figure 2 It is a schematic diagram of bidirectional search in the present invention;

[0040] Figure 3 It is a schematic diagram of extended neighborhood selection in the present invention;

[0041] Figure 4 It is a schematic diagram of angle selection extended neighborhood in the present invention;

[0042] Figure 5 It is a warehouse environment diagram of the present invention;

[0043] Figure 6 It is a schematic diagram of the improved bidirectional A-Star algorithm for planning the global path in the present invention;

[0044] Figure 7 It is a schematic diagram of the improved bidirectional A-Star algorithm combined with the TEB algorithm for planning the robot path when there are no other obstacles in the present invention;

[0045] Figure 8 It is a schematic diagram of the robot speed for the improved bidirectional A-Star algorithm combined with the TEB algorithm for planning the path when there are no other obstacles in the present invention;

[0046] Figure 9 It is a schematic diagram of the improved bidirectional A-Star algorithm combined with the TEB algorithm for planning the robot path when there are static obstacles other than shelves in the present invention;

[0047] Figure 10 It is a schematic diagram of the robot speed for the improved bidirectional A-Star algorithm combined with the TEB algorithm for planning the path when there are static obstacles other than shelves in the present invention;

[0048] Figure 11Schematic diagram of the improved bidirectional A-Star algorithm combined with the TEB algorithm for planning a robot path in the presence of dynamic obstacles in the present invention;

[0049] Figure 12 Schematic diagram of the robot speed for path planning by the improved bidirectional A-Star algorithm combined with the TEB algorithm in the present invention when there are dynamic obstacles. Detailed implementation manner

[0050] This embodiment provides an intelligent warehousing robot path planning method based on TEB dynamic obstacle avoidance and an improved bidirectional A-Star algorithm. Refer to Figure 1 , including the following steps:

[0051] S1: The autonomous mobile robot builds a map based on the current environment and rasterizes it. It obtains information such as obstacles according to the map content, and marks the obstacle area and the drivable area for subsequent path planning;

[0052] S2: Initialize relevant parameters. For example, define the starting point and the ending point, and initialize the node list;

[0053] S3: Calculate the adaptive dynamic factor according to the relationship between multiple nodes, introduce it into the heuristic function, and dynamically adjust the weight of the heuristic function;

[0054] S4: Perform bidirectional search according to two target points, select the expansion node according to the angle data between the parent nodes of the two paths, and select the node with the smallest cost value for expansion until the optimal path is found;

[0055] S5: Remove redundant nodes from the planned optimal path, and then use linear interpolation to smooth the path;

[0056] S6: Provide the smoothed global path to the TEB algorithm for local path planning to achieve the effect of dynamic obstacle avoidance, and plan a path that meets the kinematic requirements and feasibility of the robot.

[0057] The operation of S1 is specifically as follows: The autonomous mobile robot collects the surrounding environment information of the warehousing area through radar and dynamically builds a map using gmapping. Rasterize the built map in a two-dimensional plane, and each grid represents a unit area. Detect and calibrate the obstacles in the environment. The obstacle area is represented by "1", and the drivable area is represented by "0". The simulated warehousing experimental map provided in the embodiment is as Figure 5 shown.

[0058] The operations of S2 are as follows: Initialize the cost, map parameters, define the target point and the starting point, and initialize two Openlist open node lists and two Closelist closed node lists respectively. Put the target point and the starting point into the Openlist1 list and the Openlist2 list respectively, corresponding to the paths path1 and path2.

[0059] The operations of S3 are as follows: In the heuristic function, the actual cost of the robot's travel is represented by the Euclidean distance, and the estimated cost is represented by the Manhattan distance. Calculating the total cost using these two distances together better meets the path requirements of the warehousing environment. Calculate the adaptive dynamic factor according to the relationship between multiple nodes and dynamically adjust the weight of the estimated cost. The specific calculation method is as follows:

[0060] (1) Select the Euclidean distance and the Manhattan distance to calculate the cost. The formula is expressed as:

[0061] f(i) = g(i) + h(i) (1)

[0062]

[0063] h(i) = |x t -x i | + |y t -y i | (3)

[0064] Among them, i represents the node currently being searched, g(i) represents the actual cost from the starting node to node i, h(i) is the estimated cost from node i to the target node, usually represented by the remaining distance, and f(i) represents the total cost of the current node. (x i , y i ) represents the coordinates of node i, (x s , y s ) represents the coordinates of the starting point, and (x t , y t ) represents the coordinates of the target point.

[0065] (2) Calculate the relevant dynamic factor coefficients according to the starting point, ending point, and the parent nodes of the two paths. The expression is:

[0066]

[0067] Define path1 as the path searched starting from the starting point, and path2 as the path searched starting from the target point. Among them, (x p1 , y p1 ) and (x p2 , y p2 ) are the search parent nodes of path1 and path2 respectively. ξ 1 and ξ5 are the distances from the starting point to the two parent nodes, ξ 3 and ξ 6 are the distances from the target point to the two parent nodes, ξ 2 is the distance between the two parent nodes, ξ 4 is the distance from the starting point to the target point. Then the improved heuristic functions for path1 and path2 are:

[0068] f 1 (i) = g 1 (i) + (0.2 * δ 1 + 0.8 * α 1 ) * h 1 (i) (5)

[0069] f 2 (i) = g 2 (i) + (0.2 * δ 2 + 0.8 * α 2 ) * h 2 (i) (6)

[0070] where δ l is determined by the ratio of the sum of ξ 1 and ξ 2 to ξ 1 , and δ 2 is calculated from the ratio of the sum of ξ 2 and ξ 3 to ξ 3 . α 1 is determined by the ratio of ξ 4 and ξ 5 , and α 2 is determined by the ratio of ξ 4 and ξ 6 .

[0071] The operation of S4 is specifically as follows: Define path1 to start searching from the starting point and path2 to start searching from the ending point, and the two paths are searched simultaneously. They are represented by the green path and the red path respectively in Figure 2 . During the search process, calculate the angle between the two paths based on the coordinates of their parent nodes, and the parent nodes select the nodes to be expanded according to the angle value. As shown in Figure 3 and Figure 4 , if the included angle range is in [0° to 90°), select the expanded nodes 1, 2, 3, 5, 8, and discard the nodes 4, 4, 7. Select the node with the minimum cost value for expansion, and repeat this step. When the two parent nodes meet as shown by the blue circle in Figure 2 , the path planning task is completed.

[0072] The operation of S5 is specifically as follows: Process the path generated in S4. The planned path in S4 usually contains redundant nodes as shown by the red dotted path in Figure 6 . Therefore, remove the redundant nodes from this path to make the path have fewer turns and be more concise, as shown by the blue dotted path in Figure 6 . The path after removing redundant nodes is not smooth enough and thus does not meet the robot kinematic constraints. Therefore, use the linear interpolation method to smooth the path, as shown by the green solid path in Figure 6 .

[0073] The operation of S6 is specifically as follows: Provide the optimized global path in S5 to the TEB algorithm. The TEB algorithm decomposes the global path into multiple local paths and optimizes the local paths according to constraints such as time optimality and kinematics. When only the storage shelves are obstacles, the robot path planned by the present invention is as shown in Figure 7 , and the speed of the robot is as shown in Figure 8 ; When there are other static obstacles in addition to the shelves, the robot path planned by the present invention is as shown in Figure 9 , and the speed of the robot is as shown in Figure 10 ; When a dynamic obstacle is encountered during the robot's driving process, the robot path planned by the present invention is as shown in Figure 11 , and the speed of the robot is as shown in Figure 12 . When an obstacle is encountered during the robot's driving process and affects its driving, adjust the local path according to the obstacle avoidance constraint conditions to avoid the dynamic obstacle and continue driving.

[0074] The above results show that an intelligent warehouse robot path planning method based on TEB dynamic obstacle avoidance and improved bidirectional A-Star algorithm proposed by the present invention can effectively plan an optimal path that meets the robot kinematics in an intelligent warehouse environment and can avoid static and dynamic obstacles during the driving process.

[0075] Of course, the above description is not a limitation of the present invention, and the present invention is not limited to the above examples. Changes, modifications, additions or substitutions made by those skilled in the art within the essence of the present invention should also fall within the protection scope of the present invention.

Claims

1. A path planning method for an intelligent warehouse robot based on TEB dynamic obstacle avoidance and improved bidirectional A-Star algorithm, characterized in that: The following steps are involved: S1: The autonomous mobile robot builds a map based on the current environment and rasterizes it, obtains information such as obstacles based on the map content, and marks the obstacle area and the drivable area for subsequent path planning; S2: Initialize related parameters, such as defining the starting point and end point, and initializing the node list; S3: Calculate the adaptive dynamic factor based on the relationship between multiple nodes, introduce it into the heuristic function, and dynamically adjust the weight of the heuristic function; S4: Perform a bidirectional search based on the two target points, select an expansion node based on the angle data between the parent nodes of the two paths, and select the node with the smallest cost value for expansion until the optimal path is found; S5: Remove redundant nodes from the planned optimal path and then use linear interpolation to smooth the path; S6: Provide the smoothed global path to the TEB algorithm to perform local path planning to achieve the effect of dynamic obstacle avoidance and plan a feasible path that meets the robot's kinematic requirements.

2. According to claim 1, a path planning method for an intelligent warehouse robot based on TEB dynamic obstacle avoidance and improved bidirectional A-Star algorithm is characterized in that: In S1, the autonomous mobile robot collects information about the surrounding environment of the storage area through radar and uses gmapping to dynamically build a map; the established map is rasterized on a two-dimensional plane, with each grid representing a unit area, and obstacles in the environment are detected and calibrated. The obstacle area is represented by "1" and the drivable area is represented by "0".

3. According to claim 1, a path planning method for an intelligent warehouse robot based on TEB dynamic obstacle avoidance and improved bidirectional A-Star algorithm is characterized in that: In S2, initialize the cost, map parameters, define the target point and the starting point, initialize two Openlist open node lists and two Closelist closed node lists respectively; put the target point and the starting point into the Openlist1 list and the Openlist2 list respectively, corresponding to the paths path1 and path2 respectively.

4. According to claim 1, a path planning method for an intelligent warehouse robot based on TEB dynamic obstacle avoidance and improved bidirectional A-Star algorithm is characterized in that: In the heuristic function of S3, the actual cost of the robot's travel is expressed by Euclidean distance, and the estimated cost is expressed by Manhattan distance. Using these two distances to calculate the total cost is more in line with the path requirements of the warehouse environment; the adaptive dynamic factor is calculated based on the relationship between multiple nodes, and the weight of the estimated cost is dynamically adjusted. The specific calculation method is as follows: (1) Select Euclidean distance and Manhattan distance to calculate the cost, and the formula is expressed as: f(i)=g(i)+h(i) (1) h(i)=|x t -x i |+|y t -y i | (3) Where i represents the node currently being searched, g(i) represents the actual cost from the starting node to node i, h(i) is the estimated cost from node i to the target node, usually expressed as the remaining distance, and f(i) represents the total cost of the current node; (x i ,y i ) represents the coordinates of node i, (x s ,y s ) represents the coordinates of the starting point, (x t ,y t ) represents the target point coordinates; (2) Calculate the relevant dynamic factor coefficients based on the starting point, end point and parent nodes of the two paths. The expression is: Define path1 as the path to search from the starting point, and path2 as the path to search from the target point; where (x p1 ,y p1 ) and (x p2 ,y p2 ) are the search parent nodes of path1 and path2 respectively; ξ1 and ξ5 are the distances from the starting point to the two parent nodes respectively, ξ3 and ξ6 are the distances from the target point to the two parent nodes respectively, ξ2 is the distance between the two parent nodes, and ξ4 is the distance from the starting point to the target point; then the improved heuristic functions of path1 and path2 are: f1(i)=g1(i)+(0.2*δ1+0.8*α1)*h1(i) (5) f2(i)=g2(i)+(0.2*δ2+0.8*α2)*h2(i) (6)where δ1 is determined by the ratio of the sum of ξ1 and ξ2 to ξ1, and δ2 is calculated by the ratio of the sum of ξ2 and ξ3 to ξ3; α1 is determined by the ratio of ξ4 and ξ5, and α2 is determined by the ratio of ξ4 and ξ6.

5. According to claim 1, a path planning method for an intelligent warehouse robot based on TEB dynamic obstacle avoidance and improved bidirectional A-Star algorithm is characterized in that: In S4, path1 starts searching from the starting point, and path2 starts searching from the end point, and the two paths are searched simultaneously. During the search, the angle between the two paths is calculated based on the coordinates of their parent nodes, and the parent nodes of the two paths select the node to be expanded based on the angle value. The node with the smallest cost is selected for expansion, and this step is repeated. When the two parent nodes meet, the path planning task is completed.

6. According to claim 1, a path planning method for an intelligent warehouse robot based on TEB dynamic obstacle avoidance and improved bidirectional A-Star algorithm is characterized in that: The path generated in S4 is processed in S5; the planned path in S4 usually contains redundant nodes, so the redundant nodes are removed from the path to reduce turns and make the path more concise; the path without redundant nodes is not smooth enough and therefore does not meet the robot kinematic constraints, so the linear interpolation method is used to smooth the path.

7. According to claim 1, a path planning method for an intelligent warehouse robot based on TEB dynamic obstacle avoidance and improved bidirectional A-Star algorithm is characterized in that: In S6, the global path optimized in S5 is provided to the TEB algorithm; the TEB algorithm decomposes the global path into multiple local paths, and optimizes the local paths according to time optimization, kinematics and other constraints; when the robot encounters a dynamic obstacle during its driving process and affects its driving, the local path is adjusted according to the obstacle avoidance constraint conditions, so as to avoid the dynamic obstacle and continue driving.

Images