Path planning method for four-way vehicle dispatching

Abstract

The four-way vehicle scheduling path planning method described in the present invention proposes a vehicle path planning obstacle avoidance algorithm based on pre-application and occupation of points. It rationally implements obstacle avoidance management based on the real-time calculation results of the vehicle queue loop traversal, establishes point and space protection for each four-way vehicle, thereby achieving the design goals of simultaneous scheduling of multiple vehicles and rationally and efficiently completing the transportation task within the warehouse. The four-way vehicle scheduling path planning method is to pre-application and occupation of points for each running four-way vehicle every 50 to 150ms, generate the occupied nodes and brake nodes specified in this calculation; each vehicle runs according to the nodes specified in each path planning process, and the nodes that have been occupied and passed are released; before generating a new occupiable node, the vehicle stops at the brake node specified in this calculation until the four-way vehicle reaches the specified end point of the task.

Description

Technical Field

[0001] The invention relates to a four-way vehicle dispatching path planning method applied to cargo transportation in a stereoscopic warehouse, belonging to the field of logistics warehousing. Background Art

[0002] Automation and intelligent control technologies are rapidly developing in the logistics and warehousing sectors. With the increasing demand for industrial and commercial land and labor costs, intensive transit warehouses are becoming more efficient due to their ability to fully utilize space and reduce labor requirements. With the increasing adoption of high-bay warehouses, four-way shuttles, capable of integrated in-warehouse transport and verification operations, meet the inbound and outbound efficiency requirements of intensive storage and have become a widely used transport vehicle.

[0003] In the existing high-bay warehouses, there are usually hundreds of four-way vehicles configured and used. The upper-level WMS warehouse management system issues several transportation tasks in batches, and completes various in-and-out warehouse operations based on scheduling instructions and reasonable route selection. How to achieve high-efficiency operation of four-way vehicles and optimize scheduling accuracy is the core of cargo management and transportation in high-bay warehouses. The four-way vehicles currently used cannot directly receive tasks from upper-level systems such as WCS, and lack systematic and efficient global calculations in route planning, which often leads to problems such as rigid routes and deadlocks in multiple deep storage locations. It is impossible to randomly change the optimal route and the most suitable vehicle, so the operating efficiency is low and the in-and-out time is long. It is often even necessary to manually intervene in the operation of the vehicle, and it is impossible to truly realize the operation and management of unmanned warehouses.

[0004] In view of this, this patent application is filed. Summary of the Invention

[0005] The four-way vehicle dispatching path planning method proposed in this application aims to solve the problems existing in the above-mentioned existing technologies and proposes a vehicle path planning obstacle avoidance algorithm based on the pre-application and occupancy of points. The obstacle avoidance management is reasonably implemented with the real-time calculation results of the vehicle queue cyclic traversal, and the point and space protection for each four-way vehicle is established, thereby achieving the design purpose of simultaneous dispatching of multiple vehicles and reasonably and efficiently completing the transportation tasks within the warehouse.

[0006] To achieve the above design objectives, the four-way vehicle scheduling path planning method is to pre-apply and occupy points for the running four-way vehicles one by one every 50 to 150ms, and generate the occupied nodes and braking nodes specified in this calculation; each vehicle runs according to the nodes specified for occupation in each path planning process, and the occupied and passed nodes are released; before generating a new occupiable node, the vehicle stops at the braking node specified in the last calculation until the four-way vehicle reaches the specified end point of the task.

[0007] Furthermore, the four-way vehicle dispatching path planning method includes the following execution steps:

[0008] 1) Clear the memory data of the previous stage obstacle avoidance algorithm;

[0009] Using memory for data calculation can effectively improve the calculation speed and avoid the time loss caused by reading the database;

[0010] 2) Read the running status data of all four-way vehicles and put the vehicles that meet the following conditions into the queue for node reservation calculation;

[0011] 2.1) Determine whether the vehicle is assigned a path containing an occupiable node. Vehicles that are not assigned a path are not counted.

[0012] 2.2) Determine the vehicle's operating status and do not implement path planning for vehicles that are not operating normally;

[0013] 3) According to the destination of the dispatch task, apply for a node for each vehicle in the order of the queue;

[0014] 3.1) Read the current vehicle's operating data, including the path nodes specified in the previous stage, and find the node that is on the same line as the current point of the current vehicle and is farthest away;

[0015] 3.2) Calculate the nodes that can be applied for this time;

[0016] i) Calculate the occupiable nodes specified in the previous calculation stage;

[0017] ii) Calculate the number of nodes that can be applied for in this stage

[0018] Read the configuration data of the current vehicle, determine the fixed length configuration value of its node application, and clearly define the number of fixed application nodes as S2;

[0019] Read the configuration data of the current vehicle, determine the farthest length configuration value of its node application, and clearly define the farthest application node number as S3;

[0020] The number of nodes from the braking point in the previous stage to the node farthest from the current node on the same straight line is S1;

[0021] The minimum of the above three node numbers is taken as the node set S applied for this stage, S = min(S1, S2, S3);

[0022] 3.3) Traverse the set S to make security judgments to generate nodes that can be occupied

[0023] i) Start traversing from the node closest to the current point of the current vehicle, and stop running when an abnormal point is detected until the entire node set S is traversed;

[0024] ii) Make safety assessments;

[0025] iii) After filtering out unsafe nodes, complete the node application and take all safe nodes as the set list of application nodes in this stage <string>;

[0026] 4) List of nodes to apply for in this stage <string>Calculate the occupied nodes for each vehicle;

[0027] 4.1) Get the collection List <string>The farthest point in

[0028] 4.2) The number of nodes between the vehicle's current point and the braking point in the previous stage is counted as the previous braking distance H1;

[0029] 4.3) Read the configuration data of the current vehicle and determine the fixed protection distance value occupied by its node, which is calculated as the number of fixed protection nodes H2;

[0030] 4.4) Read the configuration data of the current vehicle and determine the longest protection distance value occupied by its nodes, which is calculated as the longest protection node number H3;

[0031] 4.5) The smaller of the fixed number of protected nodes H2 and the maximum number of protected nodes H3 is taken as the occupied node length H4, that is, H4 = min(H2, H3);

[0032] 4.6) Determine the braking point of the occupied node interval

[0033] i) Take the larger value of the previous braking distance H1 and the occupied node length H4 to determine the braking point of this stage;

[0034] ii) The node set H between the vehicle’s current point and the braking point of this phase is the occupied node set;

[0035] iii) traverse the node set H to make security judgments and finally generate occupied nodes;

[0036] Start traversing from the node closest to the current point of the current vehicle, and stop running when an abnormal point is detected until all node sets H are traversed;

[0037] 5) The vehicle operates according to the occupied node set H;

[0038] The farthest node in the finally generated occupied node set H is used as the braking point for this calculation. This new braking point is sent to the vehicle, and the vehicle moves according to the occupied node set H until it reaches the designated braking point.

[0039] Furthermore, in the step 2), when generating the vehicle queue, the vehicles are queued according to the order in which they are powered on and put online, and each vehicle is traversed cyclically in the queue order to perform obstacle avoidance calculations.

[0040] Furthermore, the conditions for making a safety judgment in step 3.3) include: a. whether the node is deadlocked; b. whether a deadlock has occurred in the storage area; c. whether a deadlock exists in the extension line; d. whether there is cargo or vehicle obstruction at the next node; e. whether the vehicle needs to turn; f. whether the next node is occupied by other vehicles.

[0041] Furthermore, in step 4.6), a safety judgment is performed, and the judgment conditions are the same as those in step 3.3); after all unsafe nodes are screened out, node occupation is completed, and the occupied node set H is updated.

[0042] In summary, the advantages of the four-way vehicle dispatching path planning method are:

[0043] 1. The three-dimensional warehouse system used in this application can simultaneously monitor and manage multiple four-way vehicles, enabling optimal routing to efficiently complete inbound and outbound tasks assigned in batches by the upper-level WMS warehouse management system. After the routing is planned, vehicles can avoid each other appropriately, achieving high cargo delivery efficiency.

[0044] 2. Pre-screening the points that will eventually be occupied by the vehicle by pre-application of the points ahead of the vehicle can save a lot of calculations in the subsequent vehicle path planning algorithm and improve the calculation efficiency accordingly.

[0045] 3. This application adopts a flexible configuration of the point application length, which can better adapt to vehicles with different speeds and avoid the impact of communication time on the algorithm, thereby avoiding vehicle operation jams caused by slow algorithm result delivery; 4. This application adopts a flexible configuration of the braking distance, which can be configured with different braking distances according to different vehicles, thereby obtaining the shortest braking point and avoiding sudden braking of the vehicle or the problem of failure to stop due to a braking point that is too close.

[0046] 5. This application adopts a variety of preset safety judgment conditions to make operational safety judgments on the points that the vehicle is about to occupy, thereby ensuring the safety of the points to be occupied. Only one judgment is required, which significantly saves computing power.

[0047] 6. This application issues the braking point after occupying the node in real time during the vehicle operation, which can achieve non-stop operation under the premise of dynamically modifying the destination, thereby improving the flexibility and transportation efficiency of four-way vehicle operation. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] The present invention will now be further described with reference to the following drawings.

[0049] Figure 1 This is a flowchart of the execution of a four-way vehicle dispatching task using the path planning method described in this application;

[0050] Figure 2 It is a flowchart of the application node;

[0051] Figure 3 It is a flowchart of the application process for occupation;

[0052] Figure 4 It is a calculation diagram of the application node;

[0053] Figure 5 It is a calculation diagram for application occupation; DETAILED DESCRIPTION

[0054] Example 1, as Figure 1 As shown, the present application proposes a new four-way vehicle dispatching task execution process, which is based on the path planning method as the core, and allocates the points to be occupied for each vehicle in real time through obstacle avoidance calculation, so as to realize the uninterrupted operation of multiple four-way vehicles while ensuring the braking distance, thereby maximizing the use of route resources, completing the cargo transportation task efficiently, and realizing the unified and real-time dispatching management of multiple four-way vehicles.

[0055] The four-way vehicle dispatching task execution method comprises the following steps:

[0056] 1. The upper management system such as WCS / WMS / MES issues the pallet movement task to the four-way vehicle, generating a scheduling system task;

[0057] 2. The four-way vehicle dispatching system uses the optimized A* algorithm to select idle vehicles to execute the task according to the task content;

[0058] 3. Based on the vehicle's destination, the optimized D* algorithm is used to calculate the path, and the calculated planned route is stored in memory.

[0059] 4. Apply for and occupy the calculated path nodes, and send the mobile task to the executing vehicle in a single step;

[0060] 5. During the movement of the vehicle, the node application and occupation process for vehicle obstacle avoidance is repeatedly executed every 100ms, i.e., the four-way vehicle dispatching path planning method described in this application;

[0061] 6. The vehicle arrives at the designated destination and performs the tasks specified in the dispatch task, such as lifting or lowering the cargo;

[0062] 7. Adjust the vehicle status and feedback the real-time location to the upper management system, waiting for the issuance of new scheduling tasks.

[0063] In step 5 above, the four-way vehicle dispatching path planning method implements point pre-application and occupation for each of the four-way vehicles in operation at intervals (this time is configurable and is generally set between 50ms and 200ms), such as 100ms, to generate the occupied nodes and brake nodes specified in this calculation.

[0064] Each vehicle runs according to the nodes specified in each route planning process, and the nodes that have been occupied and passed are released;

[0065] Before generating a new occupiable node, the vehicle stops at the brake node specified in this calculation until the four-way vehicle reaches the end point specified by the task.

[0066] Specifically, if Figures 1 to 5 As shown, the four-way vehicle dispatching path planning method includes the following steps:

[0067] 1) Clear the memory data of the previous stage obstacle avoidance algorithm;

[0068] Using memory for data calculation can effectively improve the calculation speed and avoid the time loss caused by reading the database;

[0069] 2) Read the running status data of all four-way vehicles and put the vehicles that meet the following conditions into the queue for node reservation calculation;

[0070] 2.1) Determine whether the vehicle is assigned a path containing an occupiable node. Vehicles that are not assigned a path are not counted.

[0071] 2.2) Determine the vehicle's operating status and do not implement path planning for vehicles that are not operating normally;

[0072] When generating a vehicle queue, it is preferred to queue each vehicle in the order in which it is powered on and online, and then traverse each vehicle in the queue order to perform obstacle avoidance calculations.

[0073] 3) Assuming the destination of the scheduling task is point C, apply for a node for each vehicle in the order of the queue;

[0074] 3.1) Read the current vehicle's operating data, including the path nodes specified in the previous stage, and find the node (such as point B) that is on the same line as the current point of the current vehicle (such as point A) and is farthest away;

[0075] Since the vehicle is constantly in motion, its real-time position is constantly changing. To determine the farthest node on the same straight line, we traverse all nodes in reverse order, starting from point C, the endpoint of the scheduling task. We compare the vector angle between each node and the vehicle's current node. Nodes whose angles exceed the system's default tolerance value are the nodes where the vehicle needs to turn. Otherwise, the node is on the same straight line as the current point.

[0076] The reason for searching for the farthest node among the nodes on the same line as the current point is that each time a four-way vehicle turns, it spends about 2 to 3 seconds waiting for the hydraulic components to execute the turning action. This waiting time is significantly longer than the 100ms interval between each path planning cycle described in this application. Therefore, the node generated and sent to the vehicle for execution each time can point to the turning point. When the vehicle turns, the obstacle avoidance calculation for the next straight line segment can be performed.

[0077] 3.2) Calculate the nodes that can be applied this time

[0078] The nodes that can be applied for include the following options:

[0079] i) Calculate the number of nodes between points A and D for the occupiable nodes specified in the previous calculation process, and obtain the number S0;

[0080] The node from the current point A to the braking point D in the previous stage is safe because the node between the vehicle and the braking point D is already occupied by the vehicle. Therefore, the node application in this stage is calculated directly from the braking point D in the previous stage, that is, the safety of the node in the current stage is calculated.

[0081] ii) Calculate the number of nodes that can be applied for in this stage

[0082] Read the current vehicle's configuration data, determine the fixed length configuration value of its node application, and specify the number of fixed application nodes as S2; the fixed length value applied by each vehicle is configured in the XML file, and the length m1 applied by the vehicle each time is obtained based on the configuration item. The number of nodes included in the process of applying length m1 from the braking point D in the previous stage is S2;

[0083] Read the current vehicle's configuration data, determine the maximum length configuration value requested by its nodes, and specify the number of nodes requested at the maximum distance as S3. The maximum length value requested by each vehicle is configured in the XML file. Based on the configuration item, the maximum length requested by the vehicle each time is m2. The number of nodes involved in the process of requesting length m2 from the braking point D in the previous stage is S3.

[0084] The number of nodes from the braking point D in the previous stage to the node B that is on the same straight line as the current node and is farthest away is S1;

[0085] The minimum of the above three node numbers is taken as the node set S applied for this stage, that is, S = min(S1, S2, S3);

[0086] The one with the smallest value among the above three node numbers S1, S2, and S3 is based on the principle of application and occupation of this application, which is to occupy the least points to leave more route options for other vehicles while ensuring the normal operation of the vehicle.

[0087] 3.3) Traverse the set S to make security judgments to generate nodes that can be occupied

[0088] i) Start traversing from the node closest to the current point of the current vehicle (point A), and stop running if an abnormal point is detected until the entire node set S is traversed;

[0089] ii) The conditions for making safety judgments are as follows;

[0090] a. Whether the node is deadlocked, that is, whether the node blocks other vehicles at the corner of the path, so as to avoid the deadlock problem of vehicles;

[0091] b. Whether a deadlock has occurred within the storage area. That is, if the next node is a storage location in the storage area with directional restrictions for entry and exit, it is necessary to determine at this node whether there is a deadlock with other vehicles entering or exiting the storage location.

[0092] c. Whether there is a deadlock on the extension line. That is, if the current node is the turning point (such as point B) that the vehicle arrives at, will the vehicle deadlock other vehicles when turning and entering the next road segment (extension line)?

[0093] d. Whether there is any cargo or vehicle obstructing the next node, including if the next node is on the reversing route, whether there is a vehicle that needs to reverse and whether there is reversing space;

[0094] e. Whether the vehicle needs to turn;

[0095] f. Whether the next node is occupied by another vehicle, that is, whether another vehicle has already applied for and occupied the node in advance, then no other vehicle may apply;

[0096] iii) Complete node application after screening out unsafe nodes

[0097] If all the above conditions are met, the node set S is taken as the set List of nodes applied for in this stage. <string>;

[0098] If one of the above conditions is not met, the node before the unsafe point is taken as the final application point, and the set List of application nodes for this stage is updated and generated. <string>;

[0099] Combine the current vehicle with the collection List <string>Binding,stored in memory implements the following occupancy calculation;

[0100] 4) List of nodes to apply for in this stage <string>Calculate the occupied nodes for each vehicle;

[0101] The occupied node refers to the range where the current vehicle operation is absolutely safe. Therefore, it needs to be calculated based on the current position of the vehicle to avoid unsafe operation caused by manual vehicle movement, vehicle shutdown, or other points going online.

[0102] 4.1) Get the collection List <string>The farthest point in (such as point E);

[0103] 4.2) The number of nodes between the vehicle's current point (point A) and the braking point D in the previous stage is counted as the previous braking distance H1;

[0104] 4.3) Read the current vehicle's configuration data and determine the fixed protection distance value occupied by its nodes, which is calculated as the number of fixed protection nodes H2. The fixed protection distance value occupied by each vehicle is configured in the XML file. Based on the configuration item, the fixed occupancy length n1 of each vehicle is obtained. The number of nodes included in the process of occupying the length n1 from the current point forward is H2.

[0105] 4.4) Read the current vehicle's configuration data and determine the maximum protection distance value occupied by its nodes, which is calculated as the maximum number of protection nodes H3. The maximum protection distance value occupied by each vehicle is configured in the XML file. Based on the configuration item, the maximum length occupied by the vehicle at each time is n2. The number of nodes included in the process of occupying length n2 from the current point forward is H3.

[0106] 4.5) The smaller of the fixed number of guard nodes H2 and the maximum number of guard nodes H3 is taken as the occupied node length H4, i.e., H4 = min(H2, H3). Here, the shortest node distance is taken to ensure that the current vehicle does not decelerate, leaving space for other vehicles to operate safely.

[0107] 4.6) Determine the braking point of the occupied node interval

[0108] i) Take the larger value of the previous braking distance H1 and the occupied node length H4 to determine the braking point of this stage;

[0109] ii) The node set H between the vehicle's current point (point A) and the braking point in this stage is the occupied node set;

[0110] iii) traverse the node set H to make security judgments and finally generate occupied nodes;

[0111] Start traversing from the node closest to the current point of the current vehicle (point A), and stop running when an abnormal point is detected until all node sets H are traversed;

[0112] The conditions for safety judgment are the same as those in step 3.3) and will not be repeated here;

[0113] After filtering out all unsafe nodes, complete node occupation and update the occupied node set H;

[0114] 5) The vehicle operates according to the occupied node set H;

[0115] The farthest node in the finally generated occupied node set H is used as the braking point for this calculation. This new braking point is sent to the vehicle, and the vehicle moves according to the occupied node set H until it reaches the designated braking point.

[0116] During the above operation process, the path planning method described in this application can continuously send the occupied node set H to each vehicle, and the vehicle can continuously correct its braking point and destination address, thereby ensuring that the vehicle will not stop and continue to run to the end point.

[0117] The embodiments described above, in conjunction with the accompanying drawings, are merely preferred solutions for achieving the objectives of the present invention. Those skilled in the art will readily be able to derive alternative structures consistent with the design concepts of the present invention based on these insights. Other structural features derived from these alternatives are also intended to fall within the scope of the present invention.< / string> < / string> < / string> < / string> < / string> < / string> < / string> < / string>

Claims

1. A four-way vehicle dispatching path planning method, characterized by: Every 50 to 150 ms, the points are pre-application and occupied one by one for the four-way vehicles in operation, and the occupied nodes and brake nodes specified for this calculation are generated; Each vehicle runs according to the nodes specified in each route planning process, and the nodes that have been occupied and passed are released; Before generating a new occupiable node, the vehicle stops at the brake node specified in the last calculation until the four-way vehicle reaches the end point specified by the task; The following steps are included: 1) Clear the memory data of the previous stage obstacle avoidance algorithm; 2) Read the running status data of all four-way vehicles and put the vehicles that meet the following conditions into the queue for node reservation calculation; 2.1) Determine whether the vehicle is assigned a path containing an occupiable node. Vehicles that are not assigned a path are not counted. 2.2) Determine the vehicle's operating status and do not implement path planning for vehicles that are not operating normally; 3) According to the destination of the dispatch task, apply for a node for each vehicle in the order of the queue; 3.1) Read the current vehicle's operating data, including the path nodes specified in the previous stage, and find the node that is on the same line as the current point of the current vehicle and is farthest away; 3.2) Calculate the nodes that can be applied for this time; i) Calculate the occupiable nodes specified in the previous calculation stage; ii) Calculate the number of nodes that can be applied for in this phase; Read the configuration data of the current vehicle, determine the fixed-length configuration value requested by its node, and identify the set of fixed-length request nodes and count them as S2; Read the configuration data of the current vehicle, determine the farthest length configuration value requested by its node, and identify the set of farthest requesting nodes and count them as S3; The set of nodes from the braking point in the previous stage to the node that is on the same straight line as the current node and is farthest away is counted as S1; The set with the least number of nodes among the above three sets is taken as S, S = min(S1, S2, S3); 3.3) Traverse the set S to make safety judgments to generate occupiable nodes; i) Start traversing from the node closest to the current point of the current vehicle, and stop running when an abnormal point is detected until the entire node set S is traversed; ii) Make safety assessments; iii) After filtering out unsafe nodes, complete the node application and take all safe nodes as the set list of nodes applied for this stage <string> ；< / string> 4) List of nodes to apply for in this stage <string> Calculate the occupied nodes for each vehicle;< / string> 4.1) Get the collection List <string> The farthest point in< / string> 4.2) The distance between the vehicle's current point and the braking point in the previous stage is counted as the previous braking distance H1; 4.3) Read the configuration data of the current vehicle and determine the fixed protection distance occupied by its node, which is calculated as the fixed protection node distance H2; 4.4) Read the configuration data of the current vehicle and determine the longest protection distance occupied by its nodes, which is calculated as the longest protection node distance H3; 4.5) The smaller of the fixed protection node distance H2 and the longest protection node distance H3 is calculated as the occupied node length H4, that is, H4 = min(H2, H3); 4.6) Determine the braking point of the occupied node interval; i) Take the farthest point pointed by the previous braking distance H1 and the occupied node length H4 and determine it as the braking point of this stage; ii) Take the node set between the vehicle’s current point and the braking point of this phase and count it as the occupied node set H; iii) traverse the occupied node set H to perform security judgment to finally generate the occupied nodes; Start traversing from the node closest to the current point of the current vehicle, and stop running when an abnormal point is detected until all occupied node sets H are traversed; 5) The vehicle operates according to the occupied node set H; The farthest node in the finally generated occupied node set H is used as the braking point for this calculation. This new braking point is sent to the vehicle, and the vehicle moves according to the occupied node set H until it reaches the designated braking point.

2. The four-way vehicle dispatching path planning method according to claim 1, characterized in that: In the step 2), when generating the vehicle queue, the vehicles are queued according to the order in which they are powered on and put online, and each vehicle is traversed cyclically in the queue order to perform obstacle avoidance calculations.

3. The four-way vehicle dispatching path planning method according to claim 1, characterized in that: The conditions for performing security judgment in step 3.3) include: a. whether the node is deadlocked; b. whether a deadlock has occurred in the storage area; c. Whether there is deadlock in the extension line; d. Is there any cargo or vehicle blocking the next node? e. Does the vehicle need to turn? f. Whether the next node is occupied by other vehicles.

4. The four-way vehicle dispatching path planning method according to claim 1, characterized in that: In step 4.6), a safety judgment is performed, and the judgment conditions are the same as those in step 3.3); After filtering out all unsafe nodes, node occupation is completed and the occupied node set H is updated.

Images