CN116205474B

CN116205474B - AGV task allocation method and device for parking lot, electronic equipment and storage medium

Info

Publication number: CN116205474B
Application number: CN202310501691.2A
Authority: CN
Inventors: 牛伟明; 周皓然; 叶绍泽; 苗东菁; 陆国锋; 黎治华; 李雨桐; 袁杰遵; 余齐
Original assignee: Shenzhen Senge Data Technology Co ltd
Current assignee: Shenzhen Senge Data Technology Co ltd
Priority date: 2023-05-06
Filing date: 2023-05-06
Publication date: 2023-07-18
Anticipated expiration: 2043-05-06
Also published as: CN116205474A

Abstract

The invention discloses an AGV task allocation method, an AGV task allocation device, electronic equipment and a storage medium of a parking lot, wherein a scheduling allocation strategy for tasks is output by inputting an AGV vehicle set and a task list of the parking lot, so that different parking tasks are automatically allocated to a plurality of AGVs; secondly, a group of optimal task allocation schemes are obtained through repeated scheme recombination, and an objective function for balancing the operation cost and the parking efficiency of the parking lot is constructed at the same time; then, the objective function can be utilized to calculate the scheme matching degree of each scheme in the optimal group of task allocation schemes, and finally, the task allocation scheme with the highest scheme matching degree can be used as the optimal task allocation scheme of the parking lot; therefore, the invention not only can realize task allocation of a plurality of AGVs at the same time, but also balances the parking operation cost and the execution task efficiency of the AGVs in the task allocation process, thus optimizing the dispatching cost of the parking lot on the basis of ensuring the execution efficiency.

Description

AGV task allocation method and device for parking lot, electronic equipment and storage medium

Technical Field

The invention belongs to the technical field of AGV task allocation, and particularly relates to an AGV task allocation method and device for a parking lot, electronic equipment and a storage medium.

Background

With the continuous increase of the quantity of the automobile, the problems of difficult and time-consuming parking are more and more serious; therefore, it is more and more important to reduce the ratio of the parking time in the whole travel process, and a new parking mode is needed to solve the foregoing problems.

Meanwhile, with the rapid development of computers, AGVs (Automated Guided Vehicle, automatic guided vehicles) have been gradually applied to parking lots; the AGV is a flexible and intelligent transport robot, can automatically transport vehicles to move along a preset path, has high safety and various transfer functions, is capable of efficiently and simultaneously operating the parking processes of a plurality of vehicles by high-efficiency cooperative operation and avoidance of uncertain and inaccurate human interference based on an intelligent unmanned parking lot of the AGV, reduces the burden of a driver, reduces the parking time consumption and becomes a novel intelligent parking mode.

At present, most of task allocation of AGVs in a parking lot is mainly single AGVs, the situation that a plurality of AGVs execute tasks is not considered, tasks cannot be allocated to the plurality of AGVs at the same time, and parking operation cost and execution task efficiency of the AGVs are not considered when task allocation is carried out; therefore, the obtained task allocation scheme cannot optimize the dispatching cost of the parking lot on the basis of ensuring the execution efficiency; based on this, how to provide a task allocation method that can simultaneously implement task allocation of a plurality of AGVs and balance execution efficiency and operation cost during task allocation has become a problem to be solved.

Disclosure of Invention

The invention aims to provide an AGV task allocation method, an AGV task allocation device, electronic equipment and a storage medium for a parking lot, which are used for solving the problems that tasks cannot be allocated to a plurality of AGVs simultaneously in the prior art, and the parking operation cost and the execution task efficiency of the AGVs are not considered in task allocation, so that the dispatching cost of the parking lot cannot be optimized on the basis of guaranteeing the execution efficiency.

In order to achieve the above purpose, the present invention adopts the following technical scheme:

in a first aspect, an AGV task allocation method for a parking lot is provided, including:

acquiring a task list and an AGV vehicle set of a target parking lot, and generating an initial task scheme set based on the task list and the AGV vehicle set, wherein the initial task scheme set comprises a plurality of initial task allocation schemes, and any initial task allocation scheme comprises parking tasks allocated to each AGV vehicle in the AGV vehicle set;

constructing a task objective function by utilizing the task list and the AGV vehicle set, wherein the task objective function is used for representing the total dispatching cost of the target parking lot, and the total dispatching cost comprises the operation cost of the target parking lot and the parking efficiency cost of the target parking lot;

Carrying out scheme recombination processing on each initial task allocation scheme in the initial task scheme set to obtain a plurality of task allocation schemes, and forming a task scheme set by utilizing the plurality of task allocation schemes;

calculating the scheme matching degree of each task allocation scheme in the task scheme set according to the task objective function, wherein the higher the scheme matching degree of any task allocation scheme is, the lower the total scheduling cost corresponding to the parking scheduling of the target parking lot based on the any task allocation scheme is represented;

judging whether scheme reorganization processing of the initial task scheme set reaches a reorganization ending condition or not;

if not, updating the initial task scheme set into the task scheme set, and carrying out scheme recombination processing on each initial task allocation scheme in the initial task scheme set again until the scheme recombination processing of the initial task scheme set reaches a recombination end condition, so that the task allocation scheme with the highest scheme matching degree in the task scheme set corresponding to the recombination end condition is used as the optimal task allocation scheme of each AGV vehicle in the target parking lot.

Based on the disclosure, the method first utilizes the task list and the AGV vehicle set of the target parking lot to generate an initial task scheme set comprising a plurality of initial task allocation schemes, wherein any initial task allocation scheme comprises parking tasks allocated to each AGV vehicle; based on this, this step corresponds to the generation of multiple task schemes to achieve task assignments for different AGV vehicles.

Then, the invention can determine the optimal task allocation scheme of the target parking lot based on the initial task scheme set; specifically, a task objective function is constructed by utilizing a task list and an AGV vehicle set, wherein the task objective function is used for representing the dispatching total cost of a target parking lot, and the total cost is composed of operation cost and parking efficiency cost; thus, the method is equivalent to constructing an objective function for balancing the operation cost and the parking efficiency of the parking lot; then, the invention starts to carry out iterative scheme recombination on each initial task allocation scheme, namely, the task allocation scheme obtained in the last recombination iteration is used as initial data in the next iteration, and the iteration is continued according to the principle until stopping when the recombination ending condition is met, and based on the principle, the process is a process of optimizing the initial task allocation scheme, namely, an optimal group of task allocation schemes are found in the continuous iteration process; furthermore, the invention also needs to calculate the scheme matching degree of each task allocation scheme in the optimal group of task allocation schemes according to the task objective function, wherein the higher the scheme matching degree is, the lower the total cost of the representation scheduling is, so that the invention can balance the operation cost and the parking efficiency, and simultaneously, the total cost is minimized, and the optimal scheme is selected; and finally, when the iteration is finished, the task allocation scheme with the highest scheme matching degree can be used as the optimal task allocation scheme of each AGV vehicle in the target parking lot, so that the allocation of the parking tasks of each AGV is completed.

Through the design, the scheduling and distributing strategy for the tasks is output by inputting the AGV vehicle set and the task list of the parking lot, so that different parking tasks are automatically distributed to a plurality of AGVs; secondly, a group of optimal task allocation schemes are obtained through repeated scheme recombination, and an objective function for balancing the operation cost and the parking efficiency of the parking lot is constructed at the same time; then, the objective function can be utilized to calculate the scheme matching degree of each scheme in the optimal group of task allocation schemes, and finally, the task allocation scheme with the highest scheme matching degree can be used as the optimal task allocation scheme of each AGV vehicle in the parking lot; therefore, the invention not only can realize task allocation of a plurality of AGVs at the same time, but also balances the parking operation cost and the execution task efficiency of the AGVs in the task allocation process, so that the dispatching cost of the parking lot is optimal on the basis of ensuring the execution efficiency, and the invention is suitable for large-scale application and popularization.

In one possible design, the task list includes a number of parking tasks, any one of the parking tasks includes a parking start point and a parking end point, and any one of the AGV vehicles in the set of AGV vehicles includes a location of the any one of the AGV vehicles;

The task objective function is constructed by utilizing the task list and the AGV vehicle set, and the task objective function comprises the following steps:

constructing an operation cost function of the target parking lot based on the task list and the AGV vehicle set according to the following formula (1), and constructing a parking efficiency cost function of the target parking lot according to the following formula (2);

（1）

（2）

in the above-mentioned formula (1),representing the operating cost function of said target parking lot, < >>Representing the AGV vehicle setClose->Middle->The time taken for the respective AGV vehicle to complete the corresponding assigned parking task, and +.>According to->Position of AGV vehicle +.>The parking start points and the parking end points of the corresponding parking tasks of the AGV vehicles are calculated;

in the above-mentioned formula (2),a parking efficiency cost function representing the target parking lot;

constructing the task objective function according to the parking efficiency cost function and the operation cost function and the following formula (3);

（3）

in the above-mentioned formula (3),representing the task objective function,/->And->All represent cost coefficients.

In one possible design, the task list further includes priorities of the parking tasks, and the any initial task allocation scheme includes a plurality of task sequences, where each task sequence corresponds to an AGV vehicle, and any task sequence includes at least one parking task allocated to the corresponding AGV vehicle, each parking task in any task sequence is arranged in order of priority from high to low, and each task sequence is associated with an AGV identifier;

Correspondingly, carrying out scheme recombination processing on each initial task allocation scheme in the initial task scheme set to obtain a plurality of task allocation schemes, and forming a task scheme set by utilizing the plurality of task allocation schemes, wherein the scheme comprises the following steps:

initializing the recombination times g, and obtaining a step length coefficient and a random step length in the g-th recombination;

for a j-th initial task allocation scheme in the initial task scheme set, calculating the arrangement position of each element in the j-th initial task allocation scheme in g-th recombination based on the step length coefficient and the random step length in g-th recombination, wherein the elements in the j-th initial task allocation scheme comprise parking tasks and AGV identifiers;

rearranging each element according to the arrangement position of each element in the jth initial task allocation scheme in the g-th recombination to obtain a recombined jth initial task allocation scheme;

adding J by 1, and calculating the arrangement positions of each element in the J-th initial task allocation scheme in the g-th recombination based on the step length coefficient and the random step length in the g-th recombination until J is equal to J, so as to obtain a plurality of recombined initial task allocation schemes, wherein the initial value of J is 1, and J is the total number of the initial task allocation schemes;

Utilizing a plurality of recombined initial task allocation schemes and each initial task allocation scheme to form a recombined task scheme set;

acquiring a reorganization probability, and selecting at least one reorganization task allocation scheme from the reorganization task scheme set based on the reorganization probability to carry out probability reorganization so as to obtain a target task scheme set after probability reorganization;

performing priority repair on each target task allocation scheme in the target task scheme set to obtain a plurality of task allocation pre-schemes, wherein parking tasks in each task sequence in any one task allocation pre-scheme are arranged according to the order of priority from high to low;

calculating a reserved weight interval of each task allocation pre-scheme, and selecting J task allocation pre-schemes from a plurality of task allocation pre-schemes based on the reserved weight interval of each task allocation pre-scheme to serve as task allocation schemes in g-th recombination;

forming a task scheme set during the g-th recombination by utilizing the task allocation scheme during the g-th recombination;

the recombination end condition comprises the maximum recombination times;

correspondingly, judging whether the scheme reorganization processing of the initial task scheme set reaches the reorganization ending condition or not includes:

Judging whether the recombination times g reach the maximum recombination times or not;

if not, adding 1 to g, updating the initial task scheme set to the task scheme set at the g-th recombination, and carrying out scheme recombination on each initial task allocation scheme in the initial task scheme set again until the recombination times g reach the maximum recombination times, wherein the initial value of g is 1.

In one possible design, calculating the arrangement position of each element in the jth initial task allocation scheme at the time of the jth recombination based on the step size coefficient and the random step size at the time of the jth recombination includes:

for any element in the j-th initial task allocation scheme, calculating the arrangement position of any element in the g-th recombination according to the following formula (4) based on the step length coefficient in the g-th recombination, the random step length in the g-th recombination and the arrangement position of any element in the j-th initial task allocation scheme;

（4）

in the above-mentioned formula (4),represents the arrangement position of any one of the elements at the g-th recombination,/or->Representing the arrangement position of said any element in said j-th initial task allocation scheme,/or- >Step size coefficient at g-th recombination, < ->Representing the random step size at the g-th recombination;

wherein,,（5）

in the above-mentioned formula (5),and representing the arrangement positions of designated elements in the initial task allocation scheme with highest scheme matching degree in the initial task scheme set, wherein the designated elements are the same elements as any element in the initial task allocation scheme with highest matching degree.

In one possible design, the reorganization probability includes a first probability and a second probability, where selecting at least one reorganization task allocation scheme from the reorganization task scheme set based on the reorganization probability to perform probability reorganization, so as to obtain a target task scheme set after probability reorganization, including:

selecting at least one reorganization task allocation scheme from the reorganization task scheme set based on the first probability to serve as a target reorganization task allocation scheme;

according to the following formula (6), calculating the probability recombination position of each element in any target recombination task allocation scheme;

（6）

in the above-mentioned formula (6),representing the probability reorganization position of the x-th element in any target reorganization task allocation scheme,/or->Representing the arrangement position of the X-th element in the any target reorganization task allocation scheme, x=1, 2 >Representing local search coefficients, ++>Representing the ordering position of the same element as the x-th element in the first target scheme, +.>Representing the ordering positions of the same elements as the x-th element in a second target scheme, wherein the first target scheme and the second target scheme are different from each other and are respectively a target recombination task allocation scheme except any target recombination task allocation scheme in the recombination task scheme set;

rearranging each element in any target reorganization task allocation scheme according to the probability reorganization position of each element in the any target reorganization task allocation scheme to obtain a probability reorganization task allocation scheme corresponding to the any target reorganization task allocation scheme, and rearranging each element in all target reorganization task allocation schemes to obtain a probability reorganization task allocation scheme corresponding to each target reorganization task allocation scheme;

replacing each selected recombination task allocation scheme in the recombination task scheme set by using a plurality of probability recombination task allocation schemes to obtain a probability recombination task scheme set after replacement;

selecting a plurality of probability reorganization task allocation schemes from the probability reorganization task scheme set based on the second probability to serve as initial directional reorganization allocation schemes;

For a z-th initial directional reorganization allocation scheme in a plurality of initial directional reorganization allocation schemes, randomly selecting two elements from the z-th initial directional reorganization allocation scheme to perform position exchange so as to obtain a directional reorganization allocation scheme corresponding to the z-th initial directional reorganization allocation scheme after the position exchange;

calculating the scheme matching degree of the directional reorganization allocation scheme and the z-th initial directional reorganization allocation scheme by using the task objective function, and judging whether the scheme matching degree of the directional reorganization allocation scheme is larger than the z-th initial directional reorganization allocation scheme;

if yes, replacing the z-th initial directional reorganization allocation scheme in the probability reorganization task scheme set by using the directional reorganization allocation scheme;

and adding 1 to Z until Z is equal to Z, and completing replacement processing of all initial directional reorganization allocation schemes in the probability reorganization task scheme set to obtain the target task scheme set, wherein the initial value of Z is 1, and Z is the total number of the initial directional reorganization allocation schemes.

In one possible design, performing priority repair on each target task allocation scheme in the target task scheme set to obtain a plurality of task allocation pre-schemes, including:

Calculating out-of-order values of all target task allocation schemes, wherein the out-of-order value of any target task allocation scheme is the logarithm of parking tasks which are not ordered according to priority in any target task allocation scheme;

extracting a target task allocation scheme with an out-of-order value greater than 0 from the target task scheme set to form an out-of-order task scheme set;

for a q-th disordered task allocation scheme in the disordered task scheme set, initializing a memory list, a disordered value search set to be empty and initializing a history minimum to be infinite;

performing position exchange on target parking tasks in the q-th disordered task allocation scheme to obtain a priority repair scheme corresponding to the q-th disordered task allocation scheme, wherein the target parking tasks are two parking tasks which are not ordered according to priority in the q-th disordered task allocation scheme;

recording a priority repair scheme corresponding to the q out-of-order task allocation scheme into a priority repair set;

calculating target disorder values of all priority repair schemes in the priority repair set, and adding each target disorder value into the disorder value search set;

Extracting a minimum target disorder value in the disorder value search set, judging whether the minimum target disorder value is larger than the historical minimum value or not, and judging whether a target position exchange mode exists in the memory list or not, wherein the target position exchange mode is a position exchange mode of a priority repair scheme corresponding to the minimum target disorder value;

if not, updating the history minimum value to the minimum target disorder value, and adding the target position exchange mode to the memory list;

judging whether the history minimum value is 0;

if not, updating the q-th disordered task allocation scheme to a priority repair scheme corresponding to the smallest target disordered value in the priority repair set, and carrying out position exchange on the target parking task in the q-th disordered task allocation scheme again until the history minimum value is 0, so as to obtain a priority repair task allocation scheme corresponding to the q-th disordered task allocation scheme;

q is added with 1, a memory list and a disordered value search set are initialized to be empty again, and an initialization history minimum value is infinity until Q is equal to Q, so that a priority repair task allocation scheme corresponding to each disordered task allocation scheme is obtained, wherein the initial value of Q is 1, and Q is the total number of disordered task allocation schemes in the disordered task scheme set;

And forming a plurality of task allocation pre-schemes by utilizing the priority repair task allocation scheme corresponding to each disordered task allocation scheme and the target task allocation scheme with the disordered value equal to 0.

In one possible design, the method includes calculating a retention weight interval of each task allocation scheme, and selecting J task allocation schemes from the plurality of task allocation schemes based on the retention weight interval of each task allocation scheme as a task allocation scheme at g-th reorganization, including:

calculating the scheme matching degree of each task allocation pre-scheme by using the task objective function, and calculating the reservation weight of each task allocation pre-scheme based on the scheme matching degree of each task allocation pre-scheme;

for a kth task allocation pre-scheme in a plurality of task allocation pre-schemes, determining a reservation weight interval of the kth task allocation pre-scheme by utilizing a reservation weight of the kth task allocation pre-scheme and a reservation weight of the kth-1 task allocation pre-scheme, wherein the right boundary of the reservation weight interval of the kth task allocation pre-scheme is the reservation weight of the kth-1 task allocation pre-scheme, the left boundary of the reservation weight interval of the kth task allocation pre-scheme is the reservation weight of the kth task allocation pre-scheme, and the right boundary of the reservation weight interval of the first task allocation pre-scheme is 0, k=1, 2;

Extracting a task allocation pre-scheme with the highest scheme matching degree from a plurality of task allocation pre-schemes;

generating a plurality of reserved numbers, wherein the number of the reserved numbers is J-D, D is the number of task allocation pre-schemes with the maximum scheme matching degree, and the value interval of any reserved number is [0,1];

selecting J-D task allocation pre-schemes from a scheme reservation set by using a plurality of reserved numbers, wherein the scheme reservation set comprises a plurality of task allocation pre-schemes which are remained after the task allocation pre-scheme with the highest scheme matching degree is deleted in the task allocation pre-schemes, and the reserved weight interval of any selected task allocation pre-scheme corresponds to a reserved number;

and forming a task allocation scheme in the g-th recombination by utilizing a task allocation scheme with the highest scheme matching degree and the selected J-D task allocation schemes.

In a second aspect, an AGV task allocation apparatus for a parking lot is provided, including:

the system comprises an acquisition unit, a control unit and a control unit, wherein the acquisition unit is used for acquiring a task list and an AGV vehicle set of a target parking lot, and generating an initial task scheme set based on the task list and the AGV vehicle set, wherein the initial task scheme set comprises a plurality of initial task allocation schemes, and any initial task allocation scheme comprises parking tasks allocated to each AGV vehicle in the AGV vehicle set;

The function construction unit is used for constructing a task objective function by utilizing the task list and the AGV vehicle set, wherein the task objective function is used for representing the total dispatching cost of the target parking lot, and the total dispatching cost comprises the operation cost of the target parking lot and the parking efficiency cost of the target parking lot;

the scheme distribution unit is used for carrying out scheme recombination processing on each initial task distribution scheme in the initial task scheme set to obtain a plurality of task distribution schemes, and forming a task scheme set by utilizing the plurality of task distribution schemes;

the matching degree calculation unit is used for calculating the scheme matching degree of each task allocation scheme in the task scheme set according to the task objective function, wherein the higher the scheme matching degree of any task allocation scheme is, the lower the total scheduling cost corresponding to parking scheduling of the target parking lot based on the any task allocation scheme is represented;

the scheme distribution unit is used for judging whether scheme recombination processing of the initial task scheme set reaches a recombination ending condition or not;

and the scheme distribution unit is also used for updating the initial task scheme set into the task scheme set when the scheme recombination processing of the initial task scheme set does not reach the recombination ending condition, and carrying out scheme recombination processing on each initial task distribution scheme in the initial task scheme set until the scheme recombination processing of the initial task scheme set reaches the recombination ending condition, so that the task distribution scheme with the highest scheme matching degree in the corresponding task scheme set when the recombination ending condition is reached is used as the optimal task distribution scheme of each AGV vehicle in the target parking lot.

In a third aspect, another apparatus for distributing an AGV task in a parking lot is provided, taking the apparatus as an electronic device, and the apparatus includes a memory, a processor, and a transceiver, which are sequentially communicatively connected, where the memory is configured to store a computer program, the transceiver is configured to send and receive a message, and the processor is configured to read the computer program, and execute an AGV task distributing method in the parking lot as in the first aspect or any one of the first aspect may be designed.

In a fourth aspect, a storage medium is provided, on which instructions are stored, which when executed on a computer perform the method for allocating tasks of an AGV in a parking lot as in the first aspect or any one of the possible designs of the first aspect.

In a fifth aspect, there is provided a computer program product containing instructions that when run on a computer cause the computer to perform the method of AGV task allocation for a parking lot as described in the first aspect or any one of the possible designs of the first aspect.

The beneficial effects are that:

(1) According to the invention, the scheduling and distributing strategy for the tasks is output by inputting the AGV vehicle set and the task list of the parking lot, so that different parking tasks are automatically distributed to a plurality of AGVs; secondly, a group of optimal task allocation schemes are obtained through repeated scheme recombination, and an objective function for balancing the operation cost and the parking efficiency of the parking lot is constructed at the same time; then, the objective function can be utilized to calculate the scheme matching degree of each scheme in the optimal group of task allocation schemes, and finally, the task allocation scheme with the highest scheme matching degree can be used as the optimal task allocation scheme of each AGV vehicle in the parking lot; therefore, the invention not only can realize task allocation of a plurality of AGVs at the same time, but also balances the parking operation cost and the execution task efficiency of the AGVs in the task allocation process, so that the dispatching cost of the parking lot is optimal on the basis of ensuring the execution efficiency, and the invention is suitable for large-scale application and popularization.

(2) According to the method, the priority of each parking task is introduced in the scheme recombination process, and the priority is repaired during recombination, so that the parking tasks distributed for each AGV in each task distribution scheme obtained through recombination can be ensured to meet the execution sequence of the task priority; thus, the practicability of the use is improved.

Drawings

FIG. 1 is a schematic flow chart of steps of an AGV task allocation method for a parking lot according to an embodiment of the present invention;

FIG. 2 is a schematic diagram of an AGV task allocation device for a parking lot according to an embodiment of the present invention;

fig. 3 is a schematic structural diagram of an electronic device according to an embodiment of the present invention.

Detailed Description

Examples:

referring to fig. 1, the method for distributing tasks of the AGVs in the parking lot provided by the embodiment can simultaneously realize task distribution of a plurality of AGVs, and can balance the operation cost of the parking lot and the execution task efficiency of the AGVs in the task distribution process, so that the dispatching cost of the parking lot can be optimized on the basis of ensuring the execution efficiency; based on the method, the method is suitable for large-scale application and popularization in the AGV intelligent parking field; alternatively, the method may be, but not limited to, run on the task distribution side, and the example task distribution side may be, but not limited to, a personal computer (Personal comJuter, PC), tablet computer, or smart phone; it should be understood that the foregoing execution subject is not limited to the embodiments of the present application, and accordingly, the operation steps of the present method may be, but are not limited to, those shown in the following steps S1 to S6.

S1, acquiring a task list and an AGV vehicle set of a target parking lot, and generating an initial task scheme set based on the task list and the AGV vehicle set, wherein the initial task scheme set comprises a plurality of initial task allocation schemes, and any initial task allocation scheme comprises parking tasks allocated to each AGV vehicle in the AGV vehicle set; in this embodiment, the example task list may include, but is not limited to, a number of parking tasks, where any parking task includes a parking start point, a parking end point, and a priority of the any parking task (the higher the priority, the earlier the representative execution order); meanwhile, the AGV vehicle set comprises a plurality of AGV vehicles and the position of each AGV vehicle; of course, in the present embodiment, the aforementioned parking start point, parking end point, and position of the AGV vehicle may be, but are not limited to, coordinate positions.

Optionally, one of the following methods for generating the initial task solution set may be, but is not limited to, as follows:

in the first step, a task allocation constraint condition is constructed, that is, the initial task allocation scheme generated in this embodiment needs to satisfy the task allocation constraint condition, where the task allocation constraint condition may be, but is not limited to, as shown below.

Constraint 1:

（7）

in the above-mentioned formula (7),indicating that the t-th parking task in the task list TS is completed by the R-th AGV vehicle in the AGV vehicle set R +.>Representing the total number of parking tasks in the task list; thus, the meaning of the aforementioned constraint 1 is: all parking tasks in the task list should be completed by each AGV vehicle in the collection of AGV vehicles.

Constraint 2:

（8）

equation (8) above indicates that a parking task can only be completed once by one AGV vehicle.

Constraint 3:

（9）

in the above-mentioned formula (9),the load of the r-th AGV is represented, i.e., constraint 3 above is used to characterize that one AGV can only carry one vehicle at the most at the same time.

Constraint 4:

（10）

（11）

in the above-mentioned formula (10) and formula (11),indicating that the r-th AGV vehicle starts from parking start a to parking end b, wherein +.>Indicating the parking node corresponding to the parking start point a (i.e. the parking end point corresponding to the parking start point a), i.e. when ∈ ->If b is b, it indicates that the parking end point corresponding to the parking start point a is b, so +.>=1, then indicates that the r-th AGV vehicle starts from the parking start point a and reachesA parking end point b corresponding to the parking start point a; otherwise, the parking end point b corresponding to the parking start point a cannot be reached.

And a second step of: acquiring a task allocation rule, wherein the task allocation rule comprises a parking task arrangement rule, and the parking task arrangement rule is arranged according to the priority of a parking task;

and a third step of: based on the task allocation constraint conditions, a plurality of initial task allocation schemes are generated by utilizing the task list, the vehicle AGV set and the task allocation rules, so that the initial task scheme set is formed by utilizing the plurality of initial task allocation schemes.

Therefore, through the explanation, a plurality of initial task allocation schemes can be generated by utilizing the constructed task allocation constraint conditions, so that data support is provided for subsequent scheme optimization.

Further, examples of any initial task allocation scheme may include, but are not limited to: each task sequence corresponds to an AGV vehicle respectively, any task sequence comprises at least one parking task distributed for the corresponding AGV vehicle, the parking tasks in any task sequence are arranged according to the order of priority from high to low, and each task sequence is associated with an AGV identifier; in particular, the different task sequences may be separated by, but are not limited to, an AGV identifier.

The following describes, by way of example, the constitution of any one of the initial task allocation schemes:

assume that any initial task allocation scheme is: { T1, T2, T3, A1, T4, T5, A2, T6, A3}, wherein T1, T2, T3, T4, T5, and T6 represent parking tasks, T1, T2, T3 form a task sequence, A1, A2, and A3 represent AGV identifiers, i.e., T1, T2, T3 is assigned to AGV vehicle 1, T4, T5 is assigned to AGV vehicle 2, T6 is assigned to AGV vehicle 3; meanwhile, for the AGV 1, the task execution sequence is T1, T2 and T3; of course, the configuration of each of the remaining initial task allocation schemes is the same as that of the foregoing example, and will not be repeated here.

After the initial task solution set is obtained, an objective function capable of balancing the operation cost of the target parking lot and the parking execution efficiency of the AGV needs to be constructed in this embodiment, so that an optimal task allocation solution is determined based on the objective function, where the construction process of the task objective function may be, but is not limited to, as shown in the following step S2.

S2, constructing a task objective function by utilizing the task list and the AGV vehicle set, wherein the task objective function is used for representing the total dispatching cost of the target parking lot, and the total dispatching cost comprises the operation cost of the target parking lot and the parking efficiency cost of the target parking lot; in specific application, the operation cost function and the parking efficiency cost function of the target parking lot can be constructed first, and then the task target function is constructed according to the operation cost function and the parking efficiency cost function, wherein the specific construction process is as shown in the following step S21 and step S22.

S21, constructing an operation cost function of the target parking lot based on the task list and the AGV vehicle set according to the following formula (1), and constructing a parking efficiency cost function of the target parking lot according to the following formula (2).

（1）

（2）

In the above-mentioned formula (1),representing the operating cost function of said target parking lot, < >>Representing the AGV vehicle set->Middle->The AGV vehicle is finishedTo correspond to the time spent by the assigned parking task, and +.>According to->Position of AGV vehicle +.>The parking start point and the parking end point of each AGV corresponding to the parking task are calculated (namely, the distance from the position of the AGV to the parking start point is calculated according to the speed of the AGV and the distance from the parking start point to the parking end point of each assigned parking task); in the above formula (2), a ∈ ->Representing a parking efficiency cost function of the target parking lot.

Thus, the operating cost of the target parking lot represents the sum of the time required for all AGV vehicles to complete their respective parking tasks, and the parking efficiency cost represents the longest of the time required for all AGV vehicles to complete their respective parking tasks.

After the operation cost function and the parking efficiency cost function are constructed, a task objective function may be constructed based on both, as shown in step S22 below.

S22, constructing the task objective function according to the parking efficiency cost function and the operation cost function and the following formula (3).

（3）

Therefore, through the detailed explanation of the task objective function construction process, the task objective function of the embodiment is composed of the operation cost and the parking efficiency cost of the target parking lot, so that the operation cost of the parking lot and the execution task efficiency of the AGV can be balanced by using the task objective function, and the dispatching cost of the parking lot is optimized on the basis of guaranteeing the execution efficiency.

After the task objective function is constructed, the task objective function may be used to perform continuous iterative update on the initial task solution set, so as to determine an optimal task solution set by using an iterative algorithm, where the iterative process is shown in the following steps S3 to S6.

S3, carrying out scheme recombination processing on each initial task allocation scheme in the initial task scheme set to obtain a plurality of task allocation schemes, and forming a task scheme set by utilizing the plurality of task allocation schemes; in this embodiment, in the process of one iteration reorganization, the arrangement position of each element in each initial task allocation scheme needs to be transformed (the element includes a parking task and an AGV identifier); then carrying out probability recombination; then, repairing the priority, and finally, determining a retainable scheme according to a task objective function so as to form a task scheme set; specifically, the recombination process may be, but is not limited to, those shown in the following steps S31 to S39.

S31, initializing the recombination times g, and obtaining a step length coefficient and a random step length in the g-th recombination; in this embodiment, the specific acquisition mode of the step size coefficient and the random step size at the g-th time of reorganization is described in detail in step S32 below.

S32, for a j-th initial task allocation scheme in the initial task scheme set, calculating the arrangement position of each element in the j-th initial task allocation scheme in g-th recombination based on the step length coefficient and the random step length in g-th recombination, wherein the elements in the j-th initial task allocation scheme comprise parking tasks and AGV identifiers; in this embodiment, since the calculation process of the arrangement position of each element in the jth initial task allocation scheme at the time of the jth reassembly is the same, any element is taken as an example for illustration, as shown below.

When the method is specifically applied, for any element in the jth initial task allocation scheme, the arrangement position of any element in the jth initial task allocation scheme is calculated according to the following formula (4) based on the step length coefficient in the g-th recombination, the random step length in the g-th recombination and the arrangement position of any element in the jth initial task allocation scheme.

（4）

In the above-mentioned formula (4),represents the arrangement position of any one of the elements at the g-th recombination,/or->Representing the arrangement position of said any element in said j-th initial task allocation scheme,/or->Step size coefficient at g-th recombination, < ->Represents the random step size at the g-th recombination, wherein +.>Following the distribution: />。

Still further, the method comprises the steps of,calculated by the following formula (5)And (3) out:

（5）

Meanwhile, on the basis of the foregoing examples, the formulas (4) and (5) are specifically described assuming that the j-th initial task allocation scheme is { T1, T2, T3, A1, T4, T5, A2, T6, A3}, wherein any element is assumed to be T2, then,then is 2 +.>The arrangement position corresponding to the element T2 in the initial task allocation scheme with the highest scheme matching degree in the initial task scheme set; of course, the arrangement position calculation process in the g-th rearrangement of the remaining elements is the same as that of the foregoing example, and will not be repeated here.

In this embodiment, the scheme matching degree of any initial task allocation scheme (such as the jth initial task allocation scheme) may be calculated by, but is not limited to, using the following formula (12) and formula (13):

（12）

（13）

in the above-mentioned formula (12),representing the maximum total cost of the scheduling corresponding to each initial task allocation scheme,/the total cost of the scheduling corresponding to each initial task allocation scheme>Represents the total cost of scheduling of the jth initial task allocation scheme (which can be calculated according to the formula (3)), J represents the total number of initial task allocation schemes, i.e. +.>The value of (2) is 1, 2; in the above formula (13), a ∈ ->The scheme matching degree of the jth initial task allocation scheme is expressed, and thus, based on the foregoing formula (13), it can be known that the lower the total scheduling cost of the initial task allocation scheme is, the higher the corresponding scheme matching degree is. />

From the above formula (12) and formula (13), the scheme matching degree of each initial task allocation scheme can be calculated; then, by means of the formula (4) and the formula (5), the arrangement position of each element in the jth initial task allocation scheme in the g-th recombination can be calculated; then rearranging according to the arrangement position of each element in the g-th recombination, so as to obtain a j-th initial task allocation scheme after recombination; meanwhile, according to the principle, the next initial task allocation scheme is recombined, so that a plurality of recombined initial task allocation schemes can be obtained, wherein the recombination process and the circulation process are shown in the following steps S33 and S34.

S33, rearranging each element according to the arrangement position of each element in the jth initial task allocation scheme in the g-th recombination to obtain a recombined jth initial task allocation scheme.

S34, adding 1 to J, and calculating the arrangement positions of each element in the J-th initial task allocation scheme in the g-th recombination based on the step length coefficient and the random step length in the g-th recombination until J is equal to J, so as to obtain a plurality of recombined initial task allocation schemes, wherein the initial value of J is 1, and J is the total number of the initial task allocation schemes.

Based on the steps S31-S34, the change of the ordering position of each element in each initial task allocation scheme can be realized, so that J recombined initial task allocation schemes are obtained; then, J initial task allocation schemes can be combined again to form a reorganization task scheme set for subsequent probability reorganization and priority repair, wherein the construction process of the reorganization task scheme set is as follows in step S35.

S35, forming a reorganization task scheme set by utilizing a plurality of reorganization initial task allocation schemes and each initial task allocation scheme; in this embodiment, assuming that there are 100 initial task allocation schemes in total in the initial task scheme set, after steps S31 to S34, 100 reorganized initial task allocation schemes are obtained; and then, using 100 initial task allocation schemes and 100 recombined initial task allocation schemes to form a recombined task scheme set.

After the reorganization task scheme set is obtained, probability reorganization can be performed on the internal 2J reorganization task allocation schemes, where the probability reorganization process is shown in the following step S36.

S36, acquiring a reorganization probability, and selecting at least one reorganization task allocation scheme from the reorganization task scheme set based on the reorganization probability to carry out probability reorganization so as to obtain a target task scheme set after probability reorganization; in particular implementations, the example reorganization probabilities may include, but are not limited to, a first probability for performing a first time of the reorganization of the probabilities and a second probability for performing a second time of the reorganization of the probabilities; in the present embodiment, the process of the two-time probability reorganization may be, but is not limited to, as shown in the following steps S36a to S36 i.

S36a, selecting at least one reorganization task allocation scheme from the reorganization task scheme set based on the first probability to serve as a target reorganization task allocation scheme; in this embodiment, assuming that the first probability is 20% and the total data in the reorganization task scheme set is 200%, then 40 reorganization task allocation schemes are randomly extracted to serve as target reorganization task allocation schemes; of course, when the first probability is the rest value, the extraction process of the target reorganization task allocation scheme is consistent with the foregoing example, and will not be described herein.

After extracting the corresponding target reorganization task allocation schemes from the reorganization task scheme set according to the first probability, the probability reorganization positions of all elements in each target reorganization task allocation scheme can be calculated, and for any target reorganization task allocation scheme, all elements can be rearranged according to the probability reorganization positions of all elements, so that the first probability reorganization of any target reorganization task allocation scheme is completed; in this embodiment, taking any target reorganization task allocation scheme as an example, the calculation process of the probability reorganization position of each element in the target reorganization task allocation scheme is specifically described as follows in step S36b.

S36b, calculating the probability recombination position of each element in any target recombination task allocation scheme according to the following formula (6).

（6）

In the above-mentioned formula (6),representing the probability reorganization position of the x-th element in any target reorganization task allocation scheme,/or->Representing the arrangement position of the X-th element in the any target reorganization task allocation scheme, x=1, 2>Representing local search coefficients, ++>Representing the ordering position of the same element as the x-th element in the first target scheme, +. >And representing the ordering positions of the same elements as the x-th element in the second target scheme, wherein the first target scheme and the second target scheme are different from each other and are respectively a target recombination task allocation scheme except any target recombination task allocation scheme in the recombination task scheme set.

In this embodiment, selecting any two target reorganization task allocation schemes except any one target reorganization task allocation scheme from the reorganization task scheme set; then, determining the ordering position of the element corresponding to the x-th element in the two selected target reorganization task allocation schemes, if the x-th element is assumed to be a parking task T4, determining the ordering position corresponding to T4 from the two selected target reorganization task allocation schemes, thereby obtainingAnd->The method comprises the steps of carrying out a first treatment on the surface of the Next, will be->And->Substituting the formula (6) to calculate the probability recombination position of the x-th element in any target recombination task allocation scheme.

Thus, according to the foregoing formula (6), the probability reorganization position of each element in any target reorganization task allocation scheme can be calculated, and then, according to the probability reorganization position of each element, each element in any target reorganization task allocation scheme is rearranged, so as to obtain the probability reorganization task allocation scheme corresponding to the any target reorganization task allocation scheme, where the first probability reorganization process is shown in the following step S36 c.

S36c, rearranging each element in any target reorganization task allocation scheme according to the probability reorganization position of each element in the target reorganization task allocation scheme to obtain a probability reorganization task allocation scheme corresponding to the target reorganization task allocation scheme, and rearranging each element in all target reorganization task allocation schemes to obtain a probability reorganization task allocation scheme corresponding to each target reorganization task allocation scheme; in this embodiment, an example is used to describe the present embodiment, if it is assumed that the x-th element is T4, its arrangement position before probability reorganization is 3, and the probability reorganization position is 2, then in the probability reorganization task allocation scheme, the parking task T4 is arranged in the second position; of course, the arrangement of the other elements is consistent with the foregoing examples, and will not be described herein; therefore, according to the same method, the probability reorganization positions of all elements in the rest target reorganization task allocation schemes can be calculated, and then the first probability reorganization of all the target reorganization task allocation schemes can be completed by rearranging according to the probability reorganization positions.

After the probability reorganization task allocation scheme corresponding to each target reorganization task allocation scheme is obtained, the probability reorganization task allocation scheme can be used for replacing each reorganization task allocation scheme selected from the reorganization task scheme set; in this embodiment, the scheme obtained after the first probability reorganization is used to replace the target reorganization task allocation scheme selected from the reorganization task scheme set, so as to complete the updating of the schemes in the reorganization task scheme set, as shown in the following step S36 d.

S36d, replacing each selected recombination task allocation scheme in the recombination task scheme set by using a plurality of probability recombination task allocation schemes so as to obtain a probability recombination task scheme set after replacement; in this embodiment, an example is used to illustrate the step, for example, if the reorganization task allocation scheme t1 in the reorganization task scheme set is a target reorganization task allocation scheme, and the probability reorganization task allocation scheme corresponding to the reorganization task allocation scheme t1 is t11, then t11 is used to replace t1 in the reorganization task scheme set, so as to complete updating of the scheme; of course, the replacement procedure of the remaining selected reorganization task allocation schemes is identical to the foregoing example, and will not be repeated here.

After the first probability reorganization is completed and the probability reorganization task scheme set is obtained, the probability reorganization of the second time can be performed, as shown in the following steps S36e to S36 i.

S36e, selecting a plurality of probability reorganization task allocation schemes from the probability reorganization task scheme set based on the second probability to serve as an initial directional reorganization allocation scheme; in the present embodiment, the extraction process in step S36e is the same as that in step S36a, and will not be repeated here; and after extracting a plurality of probability reorganization task allocation schemes as an initial directional reorganization allocation scheme, probability reorganization can be performed for the second time, as shown in the following steps S36f to S36 i.

S36f, randomly selecting two elements from a z-th initial directional reorganization allocation scheme in a plurality of initial directional reorganization allocation schemes for position exchange so as to obtain a directional reorganization allocation scheme corresponding to the z-th initial directional reorganization allocation scheme after the position exchange; in this embodiment, an example is used to describe step S36f, where, assuming that the z-th initial directional reorganization allocation scheme is t2= { T2, T1, T4, A1, T3, T6, A2, T5, A3}, then two elements are randomly selected from T2 for position exchange, if the selected elements are T2 and T4, then the directional reorganization allocation scheme T22 corresponding to T2 is: t22= { T4, T1, T2, A1, T3, T6, A2, T5, A3}; of course, the position exchange process of the two randomly selected elements of each initial directional reorganization allocation scheme is consistent with the foregoing examples, and will not be described again here; in addition, in the present embodiment, the element selected in step S36f may, but is not limited to, merely characterize the parking task.

After the directional reorganization allocation scheme of the z-th initial directional reorganization allocation scheme is obtained, judging whether the scheme obtained after the directional reorganization is carried out is better than the original scheme or not; in this embodiment, the scheme matching degree is used for measurement, where the measurement process is shown in the following steps S36g and S36 h.

S36g, calculating the scheme matching degree of the directional reorganization distribution scheme and the z-th initial directional reorganization distribution scheme by utilizing the task objective function, and judging whether the scheme matching degree of the directional reorganization distribution scheme is larger than that of the z-th initial directional reorganization distribution scheme; in this embodiment, the calculation process of the scheme matching degree of the directional reorganization allocation scheme and the z-th initial directional reorganization allocation scheme can refer to the calculation process of the scheme matching degree of the j-th initial task allocation scheme, that is, the calculation process using the foregoing formula (12) and the formula (13), which are not described herein.

After the scheme matching degree of the z-th initial directional reorganization allocation scheme and the scheme matching degree of the z-th initial directional reorganization allocation scheme corresponding to the directional reorganization allocation scheme are obtained, whether to allow the reservation of the scheme after directional reorganization can be judged according to the size relationship between the two scheme matching degrees, wherein the reservation process is as follows, as shown in step S36h.

S36h, if yes, replacing the z-th initial directional reorganization allocation scheme in the probability reorganization task scheme set by using the directional reorganization allocation scheme; when the method is specifically applied, the scheme matching degree of the directional reorganization allocation scheme is larger than that of the z-th initial directional reorganization allocation scheme, namely that the total cost of scheduling of the directional reorganization allocation scheme is lower than that of the z-th initial directional reorganization allocation scheme, so that the scheme obtained after directional reorganization can be judged to be more optimal, and at the moment, the z-th initial directional reorganization allocation scheme can be replaced by the directional reorganization allocation scheme, and the updating of the probability reorganization task scheme set is completed; of course, the replacement process is the same as the previous step S36d, and will not be described again.

After the directional reorganization of the z-th initial directional reorganization allocation scheme is completed, the directional reorganization process of the other initial directional reorganization allocation schemes can be performed according to the same principle, wherein the cyclic process is as shown in the following step S36i.

S36i, adding 1 to Z, and completing replacement processing of all initial directional reorganization allocation schemes in the probability reorganization task scheme set until Z is equal to Z, so as to obtain the target task scheme set, wherein the initial value of Z is 1, and Z is the total number of the initial directional reorganization allocation schemes.

Therefore, through the probability recombination process described in detail in the steps S36a to S36i, different probabilities can be utilized to sequentially conduct probability recombination twice on the recombination task scheme set, and therefore updating of partial schemes in the recombination task scheme set is completed according to the set probabilities, and the target task scheme set is obtained.

After the target task solution set is obtained, the g-th reorganization process is completed, but since the above description shows that the parking tasks have priority and the parking tasks with high priority are executed in the earlier order (i.e. the earlier the arrangement position is); therefore, in order to prevent the target task allocation schemes obtained by the reorganization in the foregoing steps S31 to S36 from not satisfying the task priorities, the present embodiment is further provided with a priority restoration process as shown in the following step S37.

S37, repairing priorities of all target task allocation schemes in the target task scheme set to obtain a plurality of task allocation schemes, wherein parking tasks in all task sequences in any task allocation scheme are arranged according to the order of the priorities from high to low; in this embodiment, a target task allocation scheme that needs to be subjected to priority repair may be determined first, and then, after the priority repair is performed on the determined target task allocation scheme, the specific process is as shown in the following steps S37a to S37 l.

S37a, calculating out-of-order values of all target task allocation schemes, wherein the out-of-order value of any target task allocation scheme is the logarithm of parking tasks which are not ordered according to priority in any target task allocation scheme; in this embodiment, an example is used to describe the calculation process of the out-of-order value, and it is assumed that any target task allocation scheme is { T2, T1, T3, A1, T4, T5, A2, T6, A3}, where the priority order of the parking tasks is: t1, T2, T3, T4, T5, and T6, then the parking tasks that are not prioritized in any target task allocation scheme are: t2, T1, T3, based on which the parking task pairs not prioritized are: t2 and T1, T1 and T3, and T1 and T3, such that the out-of-order value of any target task allocation scheme is 3; of course, the calculation process of the out-of-order value of each of the other target task allocation schemes is the same as that of the foregoing example, and will not be repeated here.

After obtaining the disorder value of each target task allocation scheme, the target task allocation scheme requiring the priority repair can be extracted, wherein the extraction process is as shown in the following step S37b.

S37b, extracting a target task allocation scheme with an out-of-order value greater than 0 from the target task scheme set to form an out-of-order task scheme set; in this embodiment, if the disorder value of any target task allocation scheme is greater than 0, it is indicated that there are parking tasks that are not arranged according to the priority in the any target task allocation scheme, so that priority repair is required; thus, the target task allocation scheme with the disorder value larger than 0 can be extracted, and thus a disorder task scheme set is formed.

After obtaining the disordered task scheme set, the priority repair of each disordered task allocation scheme in the disordered task scheme set can be performed, wherein the repair process is also performed according to a single scheme, as shown in the following steps S37 c-S37 l.

S37c, initializing a memory list and a disordered value search set to be empty and initializing a history minimum value to be infinity for a q-th disordered task allocation scheme in the disordered task scheme set; in this embodiment, the history minimum value is a history minimum disorder value; after initializing the memory list, the unordered value search set, and the history minimum, the priority repair is performed as shown in steps S37d to S37l below.

S37d, carrying out position exchange on target parking tasks in the q-th disordered task allocation scheme to obtain a priority repair scheme corresponding to the q-th disordered task allocation scheme, wherein the target parking tasks are two parking tasks which are not ordered according to priority in the q-th disordered task allocation scheme; in this embodiment, an example is taken to describe step S37d, and it is assumed that the q-th out-of-order task allocation scheme is: { T2, T1, T3, A1, T4, T5, A2, T6, A3}, wherein the parking tasks which are not ordered according to the priority are T2, T1 and T3, then two parking tasks of T2, T1 and T3 are selected randomly, then the position exchange is carried out, and if T2 and T1 are selected, then the position exchange is carried out on T2 and T1, thus obtaining the priority repair scheme { T1, T2, T3, A1, T4, T5, A2, T6, A3} of the q-th out-of-order task allocation scheme.

After the priority repair scheme corresponding to the q-th disordered task allocation scheme is obtained, the priority repair scheme may be added to the priority repair set, and the disordered values (herein named as target disordered values for distinguishing the disordered values) of each priority repair scheme in the priority repair set are calculated, so as to perform a subsequent priority repair process according to the calculated target disordered values, where the recording of the priority repair scheme and the calculation process of the corresponding disordered values thereof are shown in the following steps S37e and S37 f.

And S37e, recording a priority repair scheme corresponding to the q-th out-of-order task allocation scheme into a priority repair set.

S37f, calculating target disorder values of all priority repair schemes in the priority repair set, and adding each target disorder value into the disorder value search set; in a specific application, the method for calculating the target disorder value of each priority repair scheme can refer to the aforementioned step S37a, and will not be described herein.

Meanwhile, in the first repair process, the priority repair set is initialized to be empty, and after the step S37e, only the priority repair scheme corresponding to the q-th disordered task allocation scheme is substantially stored in the first repair process, so that the target disordered value of the priority repair scheme corresponding to the q-th disordered task allocation scheme is calculated.

After obtaining the target disorder value of each priority repair scheme, the minimum target disorder value can be extracted, and the priority recovery judgment operation is performed, as shown in the following step S37g.

S37g, extracting the minimum target disorder value in the disorder value search set, judging whether the minimum target disorder value is larger than the historical minimum value, and judging whether a target position exchange mode exists in the memory list, wherein the target position exchange mode is a position exchange mode of a priority repair scheme corresponding to the minimum target disorder value; in this embodiment, for the first priority repair of the qth out-of-order task allocation scheme, the smallest target out-of-order value in the out-of-order value search set is the target out-of-order value of the priority repair scheme corresponding to the qth out-of-order task allocation scheme, and at the same time, the history minimum value is infinity and the memory list is empty, so it can be determined that the judgment condition in step S37g is not satisfied, and therefore, the following step S37h is required.

Of course, if the minimum target disorder value is greater than the historical minimum value after multiple priority repair and the target position exchange mode exists in the memory list, deleting the target position exchange mode from the memory list, deleting the minimum target disorder value from the disorder value search set to obtain a target disorder value search set, and deleting a priority repair scheme corresponding to the minimum target disorder value from the priority repair set; then the unordered value search set in step S37g is updated to the target unordered value search set, and step S37g is re-executed.

Further, the specific operation procedure of step S37h is as follows:

s37h, if not, updating the history minimum value to the minimum target disorder value, and adding the target position exchange mode to the memory list; in this embodiment, the description is also based on the foregoing example, that is, the smallest target disorder value in the disorder value search set is the target disorder value of the priority repair scheme corresponding to the q-th disorder task allocation scheme, and the priority repair scheme of the q-th disorder task allocation scheme is { T1, T2, T3, A1, T4, T5, A2, T6, A3}, then the smallest target disorder value is 0, that is, the parking tasks in the priority repair scheme of the q-th disorder task allocation scheme are all arranged according to the priority; at this time, the corresponding target position exchange mode is the position exchange between T2 and T1, and the position exchange mode needs to be stored in the memory list; then, the following step S37i can be performed.

And S37i, judging whether the history minimum value is 0.

S37j, if not, updating the q-th disordered task allocation scheme to a priority repair scheme corresponding to the minimum target disordered value in the priority repair set, and carrying out position exchange on the target parking task in the q-th disordered task allocation scheme again until the history minimum value is 0, so as to obtain a priority repair task allocation scheme corresponding to the q-th disordered task allocation scheme; in this embodiment, if the history minimum value is not 0, it indicates that there are any parking tasks that are not arranged according to the priority in the priority repair scheme corresponding to the q-th disordered task allocation scheme, at this time, the priority repair scheme corresponding to the smallest target disordered value in the priority repair set needs to be used as the initial scheme of the next priority repair, that is, the q-th disordered task allocation scheme is updated to the priority repair scheme corresponding to the smallest target disordered value, and then, the process returns to S37d until the obtained history minimum value is 0, where the history minimum value is 0, which indicates that each parking task in the priority repair scheme corresponding to the q-th disordered task allocation scheme is arranged according to the priority, and at this time, the priority repair scheme corresponding to the q-th disordered task allocation scheme may be used as the priority repair task allocation scheme corresponding to the q-th disordered task allocation scheme.

After completing the priority repair of the q-th out-of-order task allocation scheme, the priority repair of each of the remaining out-of-order task allocation schemes can be completed in the same manner, wherein the cyclic repair process is as shown in the following step S37k.

S37k, self-adding 1 to Q, re-initializing a memory list and a disorder value search set to be empty, and obtaining a priority repair task allocation scheme corresponding to each disorder task allocation scheme when the minimum value of initialization history is infinity until Q is equal to Q, wherein the initial value of Q is 1, and Q is the total number of disorder task allocation schemes in the disorder task scheme set.

After completing the priority repair of all the out-of-order task allocation schemes based on the steps S37c to S37k, a target task allocation scheme with an out-of-order value of 0 may be combined to form a plurality of task allocation schemes, as shown in the following step S37l.

S37l, a plurality of task allocation pre-schemes are formed by utilizing a priority repair task allocation scheme corresponding to each disordered task allocation scheme and a target task allocation scheme with a disordered value equal to 0.

The priority repair of the target task allocation schemes with disorder values greater than 0 in the target task scheme set can be completed through the steps S37 a-S37 l, so that each scheme in the target task scheme set is used as a task allocation pre-scheme after the repair is completed; then, J task allocation pre-schemes are selected from the task allocation pre-schemes to be used as the task allocation schemes obtained in the g-th reorganization, wherein the selection process is as shown in the following step S38.

S38, calculating a reserved weight interval of each task allocation pre-scheme, and selecting J task allocation pre-schemes from a plurality of task allocation pre-schemes based on the reserved weight interval of each task allocation pre-scheme to serve as task allocation schemes in the g-th recombination, wherein the value of J is the total number of the initial task allocation schemes in the initial task scheme set; in this embodiment, the description is based on the foregoing example, that is, the total data amount in the reorganization task solution set is 200, so after the priority is restored, the total number of the task allocation pre-solutions is 200, and at this time, 100 task allocation pre-solutions need to be extracted from 200 task allocation pre-solutions, so as to be used as the task allocation solution in the g-th reorganization.

In this embodiment, the reserved weight interval of each task allocation pre-scheme may be calculated according to the scheme matching degree, and the process of selecting the scheme according to the reserved weight interval in the step S38 may be, but is not limited to, as shown in the following steps S38a to S38 f.

S38a, calculating scheme matching degree of each task allocation pre-scheme by using the task objective function, and calculating reservation weight of each task allocation pre-scheme based on the scheme matching degree of each task allocation pre-scheme; in this embodiment, for any task allocation pre-scheme, the scheme matching degrees of all the task allocation pre-schemes are summed first to obtain a sum of matching degrees; then dividing the scheme matching degree of any task allocation pre-scheme by the sum of the matching degrees to obtain the reservation weight of the any task allocation pre-scheme; of course, the reservation weights of the respective task allocation schemes are all between 0 and 1.

After the retention weight of each task allocation pre-scheme is obtained, a retention weight interval of each task allocation pre-scheme can be determined, as shown in step S38b below.

S38b, determining a reservation weight interval of a kth task allocation pre-scheme by using the reservation weight of the kth task allocation pre-scheme and the reservation weight of the kth-1 task allocation pre-scheme in a plurality of task allocation pre-schemes, wherein the right boundary of the reservation weight interval of the kth task allocation pre-scheme is the reservation weight of the kth-1 task allocation pre-scheme, the left boundary of the reservation weight interval of the kth task allocation pre-scheme is the reservation weight of the kth task allocation pre-scheme, and the right boundary of the reservation weight interval of the first task allocation pre-scheme is 0, k=1, 2.

In this embodiment, the left boundary of the reservation weight interval of the first task allocation pre-scheme is 0, the right boundary is W1 (the reservation weight of the first task allocation pre-scheme), the right boundary of the reservation weight interval of the second task allocation pre-scheme is W1, the corresponding left boundary is w1+w2 (W2 allocates the reservation weight of the second task allocation pre-scheme), and so on, so as to obtain the reservation weight interval of each task allocation pre-scheme.

After the reserved weight intervals of each task allocation pre-scheme are obtained, the scheme can be selected; when the method is applied specifically, a task allocation pre-scheme with the highest scheme matching degree is required to be reserved firstly; then, based on the reserved weight interval, a scheme is selected from the remaining task allocation schemes, wherein the specific selection process is as shown in the following steps S38c to S38 f.

S38c, extracting a task allocation pre-scheme with the highest scheme matching degree from a plurality of task allocation pre-schemes.

S38D, generating a plurality of reserved numbers, wherein the number of the reserved numbers is J-D, D is the number of task allocation pre-schemes with the maximum scheme matching degree, and the value interval of any reserved number is [0,1]; in this embodiment, assuming that J is 100 and the number of task allocation schemes with the greatest scheme matching degree is 1, the number of generated reservations is 99, that is, 99 schemes need to be selected from the remaining task allocation schemes by using 99 reservations, and the procedure is as shown in step S38e below.

S38e, selecting J-D task allocation pre-schemes from a scheme reservation set by using a plurality of reserved numbers, wherein the scheme reservation set comprises a plurality of task allocation pre-schemes with the highest scheme matching degree, and the remaining task allocation pre-schemes after deleting the task allocation pre-scheme in the scheme reservation set, and the reserved weight interval of any selected task allocation pre-scheme corresponds to a reserved number; in this embodiment, for any reserved number, determining a reserved weight interval to which the reserved number belongs, and then, using a task allocation pre-scheme corresponding to the reserved weight interval to which the reserved number belongs as a task allocation scheme in the g-th recombination; if any of the reserved numbers is 0.42, which is in the reserved weight interval corresponding to the task allocation pre-scheme T4, then the task allocation pre-scheme T4 is used as a task allocation scheme.

Through the step S38e, J-D task allocation pre-schemes can be selected, and then, the task allocation pre-scheme with the highest scheme matching degree is combined, so that the task allocation scheme in the g-th recombination can be formed, as shown in the following step S38f.

S38f, utilizing a task allocation pre-scheme with the highest scheme matching degree and the selected J-D task allocation pre-schemes to form a task allocation scheme in g-th recombination.

From the above steps S38a to S38f, J task allocation schemes can be selected from the 2J task allocation schemes, and the J task allocation schemes are used as the task allocation scheme in the g-th reorganization; then, the task scenario set at the g-th recombination can be obtained, as shown in step S39 below.

S39, forming a task scheme set during the g-th recombination by utilizing the task allocation scheme during the g-th recombination.

The primary scheme reorganization of the initial task scheme set can be completed through the steps S31-S39, so that a task scheme set in the g-th reorganization is obtained, at the moment, the scheme matching degree of each task allocation scheme in the task scheme set in the g-th reorganization needs to be calculated, and whether the g-th reorganization reaches the reorganization ending condition is judged; if the reorganization ending condition is reached, the reorganization process can be ended, and a task allocation scheme with the highest scheme matching degree is selected from the task scheme set in the g time reorganization, so as to serve as an optimal task allocation scheme of the target parking lot; if the recombination end condition is not reached, the task scheme set at the g-th recombination is used as the initial data at the next recombination, and the steps S31-S39 are repeated until the recombination end condition is reached; the calculation process of the scheme matching degree of each task allocation scheme in the task scheme set at the g-th recombination is shown in the following step S4.

S4, calculating the scheme matching degree of each task allocation scheme in the task scheme set according to the task objective function, wherein the higher the scheme matching degree of any task allocation scheme is, the lower the total scheduling cost corresponding to parking scheduling of the target parking lot based on the any task allocation scheme is represented; in this embodiment, step S4 is to calculate the scheme matching degree for each task allocation scheme in the task scheme set in the g-th reorganization, and the calculation process of the scheme matching degree in this step may refer to the calculation process of the scheme matching degree of the j-th initial task allocation scheme, which is not described herein.

After obtaining the scheme matching degree of each task allocation scheme in the task scheme set at the g-th recombination, the judgment of the recombination end condition can be performed as shown in the following step S5.

S5, judging whether scheme recombination processing of the initial task scheme set reaches a recombination ending condition or not; in the present embodiment, the reorganization end condition may include, but is not limited to, a maximum reorganization number and reorganization accuracy; thus, step S5 may be: judging whether the recombination times g reach the maximum recombination times or judging whether the recombination precision at the g time is not changed any more, wherein the recombination precision at the g time is the maximum scheme matching degree at the g time, and if so, the recombination precision at the g time is the same as the recombination precision at the g-1 time, then the recombination ending condition can be judged to be reached; if any of the above conditions is not satisfied, it is necessary to perform the recombination again, as shown in step S6 below.

S6, if not, updating the initial task scheme set into the task scheme set, and carrying out scheme recombination processing on each initial task allocation scheme in the initial task scheme set again until the scheme recombination processing of the initial task scheme set reaches a recombination end condition, so that a task allocation scheme with highest scheme matching degree in the corresponding task scheme set when the recombination end condition is reached is used as an optimal task allocation scheme of each AGV vehicle in the target parking lot; in this embodiment, g is self-added by 1, and the initial task scheme set is updated to the task scheme set at the g-th time of reorganization, and scheme reorganization is performed on each initial task allocation scheme in the initial task scheme set again until the reorganization times g reach the maximum reorganization times or the reorganization precision at the g-th time of reorganization is no longer changed, where the initial value of g is 1.

When it is determined that the g-th recombination does not meet any recombination condition, the task allocation set in the g-th recombination needs to be used as initial data in the next recombination, and then the step S3 is returned to perform scheme recombination again until the recombination end condition is reached; when the reorganization ending condition is reached, a task allocation scheme with highest scheme matching degree in a task scheme set corresponding to the reorganization ending condition is used as an optimal task allocation scheme of the target parking lot; if the recombination ending condition is reached in the 5 th recombination, then the task allocation scheme with the highest scheme matching degree in the task scheme set in the 5 th recombination is used as the optimal task allocation scheme; then, according to the optimal task allocation scheme, parking tasks can be allocated to each AGV vehicle.

Therefore, by the AGV task allocation method of the parking lot, which is described in detail in the steps S1 to S6, task allocation of a plurality of AGVs can be realized simultaneously, and the operation cost of the parking lot and the execution task efficiency of the AGVs can be balanced in the task allocation process, so that the dispatching cost of the parking lot can be optimized on the basis of ensuring the execution efficiency; based on the method, the method is suitable for large-scale application and popularization in the AGV intelligent parking field.

As shown in fig. 2, a second aspect of the present embodiment provides a hardware device for implementing the method for distributing tasks of an AGV in a parking lot according to the first aspect of the present embodiment, including:

The working process, working details and technical effects of the device provided in this embodiment may refer to the first aspect of the embodiment, and are not described herein again.

As shown in fig. 3, a third aspect of the present embodiment provides another apparatus for distributing tasks of an AGV in a parking lot, taking the apparatus as an electronic device, including: the AGV task allocation method of the parking lot comprises a memory, a processor and a transceiver which are sequentially communicated, wherein the memory is used for storing a computer program, the transceiver is used for receiving and transmitting messages, and the processor is used for reading the computer program and executing the AGV task allocation method of the parking lot according to the first aspect of the embodiment.

The working process, working details and technical effects of the electronic device provided in this embodiment may refer to the first aspect of the embodiment, and are not described herein again.

A fourth aspect of the present embodiment provides a storage medium storing instructions including the method for allocating an AGV task in a parking lot according to the first aspect of the present embodiment, that is, the storage medium storing instructions, when the instructions run on a computer, the method for allocating an AGV task in a parking lot according to the first aspect of the present embodiment is executed.

The working process, working details and technical effects of the storage medium provided in this embodiment may refer to the first aspect of the embodiment, and are not described herein again.

A fifth aspect of the present embodiment provides a computer program product containing instructions that when run on a computer, which may be a general purpose computer, a special purpose computer, a computer network, or other programmable device, cause the computer to perform an AGV task allocation method for a parking lot according to the first aspect of the embodiment.

Finally, it should be noted that: the foregoing description is only of the preferred embodiments of the invention and is not intended to limit the scope of the invention. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. The AGV task allocation method for the parking lot is characterized by comprising the following steps of:

if not, updating the initial task scheme set into the task scheme set, and carrying out scheme recombination processing on each initial task allocation scheme in the initial task scheme set again until the scheme recombination processing of the initial task scheme set reaches a recombination end condition, so that the task allocation scheme with the highest scheme matching degree in the task scheme set corresponding to the recombination end condition is used as the optimal task allocation scheme of each AGV vehicle in the target parking lot;

The task list further comprises priorities of all parking tasks, the any initial task allocation scheme comprises a plurality of task sequences, each task sequence corresponds to an AGV vehicle respectively, any task sequence comprises at least one parking task allocated to the corresponding AGV vehicle, all the parking tasks in any task sequence are arranged according to the order from high priority to low priority, and each task sequence is associated with an AGV identifier;

carrying out scheme recombination processing on each initial task allocation scheme in the initial task scheme set to obtain a plurality of task allocation schemes, and forming a task scheme set by utilizing the plurality of task allocation schemes, wherein the scheme comprises the following steps:

the recombination end condition comprises the maximum recombination times;

judging whether scheme reorganization processing of the initial task scheme set reaches a reorganization ending condition or not, including:

2. The method of claim 1 wherein the task list includes a number of parking tasks, any one of the parking tasks includes a parking start point and a parking end point, and any one of the set of AGV vehicles includes a location of the any one of the AGV vehicles;

（1）

（2）

in the above-mentioned formula (1),representing the operating cost function of said target parking lot, < >>Representing the AGV vehicle set->Middle->The time taken for the respective AGV vehicle to complete the corresponding assigned parking task, and +.>According to->Position of AGV vehicle +.>The parking start points and the parking end points of the corresponding parking tasks of the AGV vehicles are calculated;

（3）

3. The method of claim 1, wherein calculating the arrangement position of each element in the jth initial task allocation scheme at the g-th reassembly based on the step size coefficient at the g-th reassembly and a random step size comprises:

（4）

in the above-mentioned formula (4),represents the arrangement position of any one of the elements at the g-th recombination,/or->Representing the arrangement position of said any element in said j-th initial task allocation scheme,/or->Step size coefficient at g-th recombination, < ->Representing the random step size at the g-th recombination;

wherein,,（5）

4. The method of claim 1, wherein the reorganization probability comprises a first probability and a second probability, wherein selecting at least one reorganization task allocation scheme from the reorganization task scheme set based on the reorganization probability for probability reorganization to obtain a target task scheme set after probability reorganization, comprising:

（6）

in the above-mentioned formula (6),represents the probability reorganization position of the x-th element in any target reorganization task allocation scheme,representing the arrangement position of the X-th element in the any target reorganization task allocation scheme, x=1, 2>Representing local search coefficients, ++>Representing the ordering position of the same element as the x-th element in the first target scheme, +.>Representing the ordering positions of the same elements as the x-th element in a second target scheme, wherein the first target scheme and the second target scheme are different from each other and are respectively a target recombination task allocation scheme except any target recombination task allocation scheme in the recombination task scheme set;

5. The method of claim 1, wherein performing priority repair on each target task allocation scheme in the set of target task schemes to obtain a plurality of task allocation schemes, comprising:

judging whether the history minimum value is 0;

6. The method according to claim 1, wherein calculating a retention weight interval for each task allocation scheme, and selecting J task allocation schemes from among the plurality of task allocation schemes as the task allocation scheme at the g-th reorganization based on the retention weight interval for each task allocation scheme, comprises:

7. An AGV task allocation apparatus for a parking lot, comprising:

the scheme distribution unit is further used for updating the initial task scheme set into the task scheme set when the scheme recombination processing of the initial task scheme set does not reach the recombination ending condition, and re-carrying out scheme recombination processing on each initial task distribution scheme in the initial task scheme set until the scheme recombination processing of the initial task scheme set reaches the recombination ending condition, so that the task distribution scheme with the highest scheme matching degree in the corresponding task scheme set when the recombination ending condition is reached is used as the optimal task distribution scheme of each AGV vehicle in the target parking lot;

the recombination end condition comprises the maximum recombination times;

8. An electronic device, comprising: the AGV task allocation method for the parking lot according to any one of claims 1 to 6, comprising a memory, a processor and a transceiver which are connected in sequence in communication, wherein the memory is used for storing a computer program, the transceiver is used for receiving and transmitting messages, and the processor is used for reading the computer program and executing the AGV task allocation method for the parking lot according to any one of claims 1 to 6.

9. A storage medium having stored thereon instructions which, when executed on a computer, perform the method of distributing an AGV task in a parking lot according to any one of claims 1 to 6.