Disclosure of Invention
The purpose of the invention is as follows: in order to solve the problems, a method for realizing dense library equipment path planning based on a dynamic edge weight topological graph is provided, model attributes are efficiently and accurately maintained by using a uniform path-finding algorithm and a path attribute management scheme, so that the algorithm is more universal, a system can be constructed and maintained in a standardized mode, model information is described by using richer attribute data in addition in various heterogeneous customized developments of a warehouse, the model can be matched with various higher-level path optimization algorithms, and the scheduling performance of the system is further improved.
The technical scheme is as follows: in order to achieve the purpose, the invention can adopt the following technical scheme:
a method for realizing dense library equipment path planning based on a dynamic edge weight topological graph comprises the following steps:
(1) establishing a topological graph model of the warehouse, and setting path weights of the topological graph, wherein the topological graph model comprises paths, path attribute information and equipment action information executed by the paths; defining the path weight attribute by adopting passing time, and obtaining an average value as an initialized parameter by performing sampling for multiple times through the passing time;
(2) planning a cargo transportation task, firstly confirming the position of a cargo, confirming the starting point and the end point of a preset path of the cargo, and carrying out various path sniffing based on a topological graph, wherein the sniffing scheme is as follows: searching all reachable points with the distance between the starting point and the end point within r to form a starting point set and an end point set, and calculating the shortest path of each point in the two sets by using a dijkstra shortest path algorithm; then, performing equipment execution action disassembly according to the searched path, and searching a traveling route with expected required minimum time to execute the task by calculating the time consumed by each action;
(3) after finding the traveling route, updating the use condition of the route, and continuously updating the path data in the traveling process until the task finishes the delivery of the goods or terminates the task accidentally;
wherein, the path attribute updating algorithm comprises the following steps:
before the task is executed, when the task selects a certain path, the number of the pre-occupied tasks of the path is plus 1, and a smaller pre-occupied weight is added on the path;
when a task is executed, equipment needs to occupy a path before passing through a specific path, the distance weight of the occupied path is increased by a weight M which is far larger than an initial value, when the shortest path is calculated by other tasks next time, if a communication channel with smaller weight of other paths exists, other channels are preferentially selected, and if a plurality of channels are occupied, a channel with the smallest sum of current weights is selected;
the equipment occupies the path section by section before advancing, when the equipment executes the action on the path and needs to occupy the path, the side weight of the path is judged to be less than M, otherwise, the occupation is carried out when the side weight is updated to be less than M;
when the equipment waits, if the waiting time exceeds a preset threshold value, a re-planning logic is triggered, namely a traveling route to a target position is re-planned by using the current position;
releasing the path every time the path passes through a section of path in the execution of the equipment, wherein the releasing method is that the edge weight value of the path at the section of path subtracts M, and the released path at the section of path can be used by other tasks;
the method for estimating the occupied time of each path comprises the following steps:
time sigma each segment path pre-occupation task number passage time of each segment path (1+ each segment fault rate)
The path with the shortest time will be planned as the execution path of the current task.
Has the advantages that: the invention uses a uniform high-abstract model to manage the whole goods conveying path of various warehouses, and uses a uniform routing algorithm and a path attribute management scheme to efficiently and accurately maintain the model attribute, so that the algorithm is more universal, and further, the system can be constructed and maintained in a standardized mode.
Detailed Description
The invention provides a method for realizing dense library equipment path planning based on a dynamic edge weight topological graph. The method is applied to a modern intelligent warehousing system and is used for controlling the traveling path of the primary and secondary shuttle vehicles or the four-way shuttle vehicle on the track.
The method comprises the following steps:
firstly, establishing a topological graph model of a warehouse, and setting a path weight of a topological graph:
the topological graph model comprises path and path attribute information and equipment action information executed by the path, wherein indexes used for calculation in the path attribute comprise the number of tasks occupied by path application, path fault rate and other auxiliary indexes besides passing time, and the numerical values of the indexes can be dynamically changed along with planning.
Path weight attributes to be used when using shortest path algorithms
The design of the path weight needs to ensure that the attribute operation conforms to the commutative law, i.e. the order of changing the attributes does not affect the final result, so the path weight attribute is defined by adopting the passing time, and the passing time (i.e. the path weight) can obtain the average value as the initialized parameter by performing sampling for multiple times
The vehicle can be segmented in the running path, the segmentation is defined according to the positioning holes or the positioning two-dimensional codes, and a section of path is formed between the two positioning holes or the two positioning two-dimensional codes.
Thereafter, planning is performed if necessary for the task
First, the position of the goods is confirmed, which is usually located at the entrance or a specific goods position, and the exception handling may be carried out on a conveying device such as a conveying line, and then the target position of the goods is requested,
after the starting point and the end point are confirmed, a plurality of possible path sniffing are carried out based on the topological graph, and the sniffing scheme is as follows: and finding all reachable points with the distances within r (which can be determined by configuration) near the starting point and the end point to form a starting point set and an end point set, and calculating the shortest path of each point in the two sets by using a dijkstra shortest path algorithm. A pruning strategy can be used in the searching process to reduce the searching workload, and if the shortest path searched before is determined to be l, all paths exceeding l +2r should be discarded. The repeated paths are merged while searching.
And then, performing equipment execution action disassembly according to the searched path, and searching a traveling route which is expected to require the least time by calculating the time consumed by each action to execute the task.
Because the data of the topological graph model is changed in real time in the task execution process, the estimated time of the planning is only used as a reference, and the deviation rectification can be carried out according to the real-time feedback condition in the actual execution process
And finally, after finding the traveling route, updating the use condition of the route, and continuously updating the path data in the traveling process until the task is finished and the goods are delivered or the route is terminated accidentally.
And the subsequent calculation can be carried out according to new data when the topological graph path attribute data is changed.
Wherein, the path attribute updating algorithm comprises the following steps:
before the task is executed, when the task selects a certain path, the number of the pre-occupied tasks of the path is plus 1, and a smaller pre-occupied weight is added on the path.
When a task is executed, equipment needs to occupy a path before passing through a specific path, and the weight of the distance between the occupied path is increased by a weight M which is far larger than the initial value, so that when the shortest path is calculated by other tasks next time, if a communication channel with smaller weight of other paths exists, other channels can be preferentially selected, and if a plurality of channels are occupied, the channel with the smallest sum of the current weights is selected
The method comprises the steps that the equipment occupies paths section by section before traveling, the length of the path occupied each time can be set by configuration, in the current algorithm, the first 3 sections of paths in the traveling direction are occupied by default, even if only one section of the path is traveled, the three sections of paths in the front can be occupied, so that two vehicles can be prevented from being collided, and meanwhile, when four-way vehicles are reversed or the traveling route needs to be changed, the circumrotation space is reserved. Meanwhile, a certain extra buffer space is reserved, so that the problem of path deadlock caused by the fact that vehicles gather near one roadway can be greatly reduced.
When the device executes the path on the path and needs to occupy the path, it needs to determine that the path edge weight is smaller than M, otherwise it needs to wait for the edge weight to be updated to be smaller than M and then can occupy,
when the device waits, such as when the waiting time exceeds 1/3 the previously estimated time or exceeds a preset threshold, the replanning logic is triggered, i.e., the traveling route to the target location is replanned using the current location.
The path is released every time when the device passes through a section of path in execution, the releasing method is that the edge weight of the path in the section is subtracted by M, the path in the section can be used by other tasks, the scheme can avoid conflict and improve the utilization rate of the path at the same time
The paths are released section by section, the problem of the minimum release unit needs to be considered, the minimum release unit can be set by configuration, in the algorithm, the default setting is that all paths except two paths behind the vehicle advancing path are released every time 6 paths are passed, so that other vehicles can be prevented from entering the road sections within 2 positioning points of the vehicle and being prevented from collision, and in addition, the problem that the rear vehicle is frequently started and stopped under the condition that the two vehicles follow to advance can be reduced through batch release paths.
The unexecuted tasks can be cancelled, the topological graph attribute can be reversely modified after cancellation, namely the number of the pre-occupied tasks is minus 1, and the increased pre-occupied weight is subtracted on the path. The same is done for task completion.
When an error occurs in the execution process, the abnormal rate of the corresponding path is updated, wherein the abnormal rate is the abnormal times/the total number of tasks executed by the path.
Estimating the occupation time formula of each path:
time sigma each segment path pre-occupation task number passage time of each segment path (1+ each segment fault rate)
The path with the shortest time is planned as the execution path of the current task
In addition, the present invention has many specific implementations and ways, and the above description is only a preferred embodiment of the present invention. It should be noted that, for those skilled in the art, without departing from the principle of the present invention, several improvements and modifications can be made, and these improvements and modifications should also be construed as the protection scope of the present invention. All the components not specified in the present embodiment can be realized by the prior art.