CN110146088B

CN110146088B - Indoor positioning navigation method and navigation model in intelligent warehouse management system

Info

Publication number: CN110146088B
Application number: CN201910523313.8A
Authority: CN
Inventors: 徐磊; 王子泰; 任远; 张红伟; 方红雨; 李晓辉
Original assignee: 安徽大学
Current assignee: Anhui Nongdao Intelligent Technology Co ltd
Priority date: 2019-06-17
Filing date: 2019-06-17
Publication date: 2020-10-27
Anticipated expiration: 2039-06-17
Also published as: CN110146088A

Abstract

The invention discloses an indoor positioning navigation method and a navigation model in an intelligent warehousing management system, wherein the navigation method comprises the following steps: collecting volume information of goods, position information of an intelligent vehicle and vacancy information of a warehouse; judging whether an intelligent vehicle in an idle state exists or not, if so, judging whether goods need to be stored or not, if so, determining a target area and a goods shelf, and determining stock vehicles; planning an inventory path and driving an inventory vehicle to travel according to the path; judging whether the stock vehicle reaches the stock point, if so, storing the target goods at the stock point; and judging whether the goods need to be taken out, if so, judging whether the goods are in the current area, and if so, taking the stock vehicle as a goods taking vehicle. Planning a goods taking path and driving a goods taking vehicle to reach a goods taking point according to the goods taking path; planning a delivery path and driving the goods taking vehicle to a delivery port. The intelligent vehicle storage system improves the storage utilization rate, can reduce the storage and taking time of the intelligent vehicle, improves the storage efficiency, and is convenient for navigating the intelligent vehicle in the storage.

Description

Indoor positioning navigation method and navigation model in intelligent warehouse management system

Technical Field

The invention relates to an indoor positioning navigation method in the technical field of warehouse management systems, in particular to an indoor positioning navigation method in an intelligent warehouse management system and an indoor positioning navigation model in the intelligent warehouse management system.

Background

In traditional storage work, manpower has inherent disadvantage in the aspect of storage transportation management, so that the operator has high labor intensity and can be tired inevitably to cause sorting and counting errors, and meanwhile, under some storage environments with potential safety hazards, such as darkness, low temperature, pollution and the like, in which flammable and explosive materials are stored, the safety of field operation is difficult to guarantee. Therefore, the intelligent three-dimensional warehouse management system has multiple functions of information processing, system control, system monitoring, system management and the like, integrates information flow and logistics, and is an important component of logistics and information flow management of modern enterprises. However, in the existing warehouse management system, the warehouse utilization rate and the warehouse efficiency during multi-point access are low, and the navigation of the intelligent vehicle in the warehouse is not facilitated.

Disclosure of Invention

Aiming at the prior technical problem, the invention provides an indoor positioning navigation method and a navigation model in an intelligent warehousing management system, which solve the problem that the warehousing utilization rate and the warehousing efficiency are low when multiple points are accessed in the prior warehousing management system.

The invention is realized by adopting the following technical scheme: an indoor positioning navigation method in an intelligent warehouse management system, which is used for navigating an intelligent vehicle for storing and taking goods in a warehouse, comprises the following steps:

step S1, collecting the volume information of the goods, the position information of the intelligent vehicle and the vacancy information of the warehouse;

step S2, judging whether an intelligent vehicle in an idle state exists; the intelligent vehicle in the idle state is defined as an idle intelligent vehicle;

if at least one idle intelligent vehicle exists, executing step S3, and judging whether goods exist at the warehousing port of the warehouse and need to be stored;

if at least one target cargo one is needed to be stored at the warehouse entry, executing step S4, determining a target area and a target shelf for the target cargo one according to the vacancy information and the volume information, and determining an idle intelligent vehicle as a stock vehicle;

a step S5 of planning an inventory path of the inventory vehicle from the inlet to the target area and the plurality of target shelves and driving the inventory vehicle to travel according to the inventory path;

step S6, determining whether the stock vehicle reaches the stock point;

executing step S7 when the stock vehicle reaches the stock point, storing the target goods in the stock point in sequence;

step S8, determining whether goods need to be taken out of the warehouse;

if a plurality of second target goods need to be taken out from the warehouse, executing step S9 to determine whether the second target goods are in the current area of the stock vehicle;

if the target cargo II is in the current area, the stock vehicle is used as a goods taking vehicle;

if the second target cargo is not in the current area, executing step S10 to determine whether there is an intelligent vehicle in an idle state; if the intelligent vehicle exists in the idle state, executing step S11, and determining an idle intelligent vehicle as the goods taking vehicle according to the position information of the goods taking point;

step S12, planning a goods taking path from the current position of the goods taking vehicle to the goods taking point, and driving the goods taking vehicle to travel to the goods taking point according to the goods taking path so as to obtain the target goods II;

step S13, planning a delivery path from the delivery point to a delivery port of the delivery vehicle, and driving the delivery vehicle to travel to the delivery port according to the delivery path;

step S14, judging whether the goods taking vehicle arrives at the goods outlet;

when the goods taking vehicle reaches the goods outlet, executing step S15, taking out the second target goods in the goods taking vehicle, and returning the goods taking vehicle to the bin entrance;

if the goods are not required to be taken out from the warehouse, executing step S16 to return the stock vehicle to the inlet;

step S17, updating the status of the stock vehicle or the delivery vehicle at the inlet to an idle status.

As a further improvement of the above scheme, the warehouse is provided with a plurality of rectangular areas, each rectangular area is provided with a plurality of roads arranged in a net shape, and two adjacent roads are intersected at one node; the number of the intelligent vehicles is multiple, each intelligent vehicle is defined as one ant, and the intelligent vehicles form an ant colony; the planning method of the stock path or the goods taking path comprises the following steps: step S501, initializing a population, a code and a genetic variable, and determining a fitness function; step S502, sequencing the individuals in the parent group according to the fitness, so that the probability of selecting the individual with higher fitness is greater than the probability of selecting the individual with lower fitness; step S503, judging whether mutation operation needs to be carried out on the ant colony; when the ant colony needs to be subjected to mutation operation, executing step S504, performing reverse mutation on the ant colony, and selecting new individuals and parent individuals according to the fitness; when the ant colony does not need to be subjected to mutation operation or the step S504 is completed, executing the step S505, and determining whether the ant colony needs to be subjected to crossover operation; when the ant colony needs to be subjected to the crossover operation, executing step S506, performing the crossover operation on the ant colony, and selecting the new individual and the parent individual according to the fitness; when the ant colony does not need to be subjected to cross operation or the step S506 is completed, executing the step S507, and judging whether the ant colony meets a preset fusion condition; when the ant colony meets the fusion condition, executing step S508, converting a solution obtained through a genetic algorithm into an initial pheromone distribution value on a path of the ant colony, and defining a path pheromone range; when the ant colony does not satisfy the fusion condition, step S502 is performed.

Further, the planning method further comprises the following steps: step S509, initializing the number of nodes, the number of ants, the cycle number and the pheromone volatilization factor that the ant colony needs to traverse, randomly placing the ants on each node, and emptying a taboo table of the ant colony; step S510, judging whether the population scale of the ant colony is not less than the maximum cycle number; when the number of loop iterations of the population is not less than the number of loop iterations, executing step S511, and outputting a search result; when the loop iteration number of the population is smaller than the loop number, step S512 is executed, and each ant selects the next node according to the following state movement rule formula:

wherein, tau_ij(t) is the pheromone track, table_kAs the tabu table, allowed_kIs a candidate set; lambda [ alpha ]_jM is the degree of urgency of the cargo_jTotal weight of cargo required for node j, d_ijThe distance from the current node to the next node; step S513, after the kth ant traverses a circle of nodes and returns to the departure point, the pheromone on the path of the kth ant is locally updated; repeating the step S512 and the step S513 until all ants traverse a circle of nodes to return to the starting point; step S514, according to the maximum fitness f_maxAnd minimum f_minUpdating the optimal path length of the iteration and the pheromones of the road sections included in the optimal path length, and the worst path of the iteration and the pheromones of the road sections included in the worst path of the iteration; step S515, reset the position of the mth ant as the starting point, and set the empty tabu table_kJudging whether the value of the information volatilization factor needs to be adjusted or not; when the value of the information volatilization factor needs to be adjusted, step S516 is executed, and after the value of the information volatilization factor ρ is adjusted according to the following formula, step S510 is executed:

when the value of the information volatilization factor does not need to be adjusted, step S510 is directly performed.

Still further, when the shortest time is used for the intelligent vehicle to stock or take goods, the following are available:

wherein f is the fitness of the fitness function;

when the inventory energy consumption of the intelligent vehicle is the lowest, the following steps are carried out:

when the intelligent vehicle has the lowest energy consumption for getting goods, the following steps are carried out:

when the intelligent vehicle inventory needs to consider cost, time and goods emergency, the following are provided:

when the intelligent vehicle gets goods and needs to consider expense, time of use and goods emergency, have:

still further, in step S508, the limiting formula of the path pheromone range is:

in step S513, the pheromone range is also limited, and the limiting formula is:

τ(r,s)←ρ·τ(r,s)+(1-ρ)·Δτ(r,s)

in step S514, the pheromone range is also updated; wherein the content of the first and second substances,

(1) the formula for updating pheromone of the optimal path length is as follows:

τ(r,s)←τ(r,s)+·Δτ(r,s)

in the formula, L_gbThe optimal path length for this iteration;

(2) the formula for performing pheromone update on the worst path length is as follows:

where (r, s) is an edge that belongs to the worst path and not to the optimal path, L_worstFor the worst path length of this path finding, L_bestThe optimal path length for the path finding is the attenuation coefficient of the pheromone on the worst path and the edge which does not belong to the optimal path after the circulation is finished, and the value range is [0, 1 ]]。

As a further improvement of the above solution, the indoor positioning navigation method further includes the following steps:

executing step S18 when the stock vehicle does not reach the stock point, adjusting the running state of the stock vehicle in real time according to the real-time position of the stock vehicle and the stock path;

and when the goods taking vehicle does not reach the goods outlet, executing step S19, and adjusting the running state of the goods taking vehicle in real time according to the real-time position of the goods taking vehicle and the goods taking path.

As a further improvement of the above solution, the method for acquiring the volume information of the cargo includes the following steps:

step S101, measuring the length L of goods positioned at the warehouse inlet; in the length direction of the goods, performing ultrasonic ranging on the goods through an ultrasonic transmitting end and starting timing; judging whether the ultrasonic ranging difference of two adjacent ultrasonic transmitting ends is greater than a first threshold value; when the ultrasonic ranging difference is larger than the threshold value one, ending timing and obtaining scanning time; calculating the product of the moving speed of the goods in the length direction and the scanning time to obtain the length L of the goods;

step S102, measuring the width W of goods at the bin entrance; by rotating one of the ultrasonic wave transmitting terminals two,generating a fan-shaped detection surface I in the width direction of the goods to perform ultrasonic ranging on the goods; judging whether the second ultrasonic ranging difference between two adjacent ultrasonic transmitting ends is larger than a second threshold value; when the second ultrasonic ranging difference is larger than the second threshold, defining the second ultrasonic transmitting end to scan the boundary of the goods; calculating the time difference delta t between the two opposite boundaries scanned by the second ultrasonic transmitting end and the goods and the distance S between the second ultrasonic transmitting end and the two boundaries_αAnd S_β(ii) a Calculating the included angle theta of the first fan-shaped detection surface_L(ii) a Wherein, theta_LV is the rotation angular velocity of the second ultrasonic wave transmitting end; calculating the width W of the cargo; wherein the content of the first and second substances,

step S103, measuring the height H of the goods at the warehouse inlet; by rotating an ultrasonic transmitting end III, a fan-shaped detection surface II is generated in the height direction of the goods so as to perform ultrasonic ranging on the goods; judging whether the ultrasonic ranging difference III of three adjacent ultrasonic ranging of the ultrasonic transmitting end is larger than a threshold value III; when the third ultrasonic transmitting end is larger than the third threshold, defining the boundary of the cargo scanned by the third ultrasonic transmitting end; calculating the distance S between the third ultrasonic transmitting end and the boundary of the goods_ψThe vertical distance S between the third ultrasonic transmitting end and the goods_ζ(ii) a Calculating the height H of the cargo:

wherein H₂Is the height of the ultrasonic transmitting end from three distances H to the ground₁Is the height of the cargo from the ground;

step S104, calculating the volume V of the cargo: v — lxwxh.

As a further improvement of the above scheme, the warehouse is provided with a plurality of rectangular areas, each rectangular area is provided with a plurality of roads arranged in a net shape, and two adjacent roads are intersected at one node; said intelligenceThe method for acquiring the position information of the vehicle comprises the following steps: s105, respectively placing two positioning modules at two ends of a diagonal line of the rectangular area, wherein the positioning modules are in wireless communication with the intelligent vehicle; step S106, calculating the distance S between the intelligent vehicle and the two positioning modules according to the communication time of the intelligent vehicle and the two positioning modules₁And S₂(ii) a Step S107, calculating the coordinates of two pre-marked points according to the coordinates of the two positioning modules; wherein, the distance between each pre-marking point and the two positioning modules is S₁And S₂(ii) a Step S108, respectively judging whether the two pre-marked points are positioned on the advancing path of the intelligent vehicle; when one of the pre-marked points is located on the traveling path and the other pre-marked point is not located on the traveling path, judging that the pre-marked point on the traveling path is the measuring position of the intelligent vehicle; and when the two pre-marked points are positioned on the advancing path, selecting the pre-marked point which does not appear on the advancing path in the previous time as the measuring position of the intelligent vehicle according to the measuring position of the intelligent vehicle in the previous time.

As a further improvement of the above scheme, the method for acquiring vacancy information of the warehouse comprises the following steps: step S109, two wireless modules respectively corresponding to the two positioning modules are respectively arranged at two ends of a diagonal line of the rectangular area; step S110, when the intelligent vehicle receives a goods taking and storing instruction, sending awakening information to the wireless module of each rectangular area so as to awaken the wireless module to acquire the position information of the intelligent vehicle through the corresponding positioning module; and step S111, driving the wireless module to sleep when the intelligent vehicle leaves the rectangular area.

The invention also provides an indoor positioning navigation model in the intelligent warehouse management system, which is used for navigating the intelligent vehicle for storing and taking goods in the warehouse and comprises a data acquisition subsystem, a data transmission subsystem and a data processing subsystem, wherein the data acquisition subsystem is used for acquiring the volume information of the goods, the position information of the intelligent vehicle and the vacancy information of the warehouse; the data transmission subsystem is used for updating the position information, the volume information and the vacancy information in real time according to the information acquired by the data acquisition subsystem; the data processing subsystem is configured to perform steps S2-S17 of any of the navigation methods described above.

The navigation method comprises the steps of collecting volume information of goods, determining a goods storage area and goods shelves according to the volume of the goods, selecting a plurality of vacant goods shelves as target goods shelves according to the volume size and the vacancy information of the goods during storage, screening a plurality of goods taking points as the target goods shelves during goods taking, determining the shortest conveying path, and navigating the intelligent vehicle by matching two-point positioning to realize multi-point access operation of an indoor storage environment so as to improve the utilization rate of storage.

In the present invention, the planning of the stock or pick path first generates an initial pheromone distribution on the problem using random search, global convergence and rapidity of the genetic algorithm. Then, the parallelism, the positive feedback mechanism and the high efficiency characteristic of the ant colony algorithm are fully utilized to carry out solving. And finally, fusion is carried out, the fused algorithm is superior to the ant colony algorithm in time efficiency and superior to the genetic algorithm in solving efficiency, and a heuristic algorithm with both time efficiency and solving efficiency is formed. In addition, the method considers the track limitation of the actual storage environment, limits the candidate set, only adds a plurality of trunk nodes which can be directly reached by the intelligent vehicle into the candidate set, improves the candidate set and further improves the solving efficiency. Meanwhile, the invention considers the combination of an ant colony algorithm and a mechanism which can effectively avoid premature convergence, improves the precision of the ant colony algorithm which can obtain the optimal performance, avoids search stagnation, strengthens the optimal solution to a greater extent, weakens the worst solution, increases the pheromone difference between the side belonging to the optimal path and the side belonging to the worst path, concentrates the search near the optimal solution more, introduces an adaptive adjustment mechanism of pheromone volatilization factors, takes the minimum path energy consumption as a main target, comprehensively considers the distribution time, and solves the distribution path which has the minimum energy consumption and short time under the condition of considering the emergency degree of goods.

The invention mainly adopts an image processing method to acquire the size of the goods in the existing warehousing management system, the acquired image information is transmitted to a processor for processing, the information amount is large, the cost is high, and the size of the goods cannot be dynamically detected.

Drawings

Fig. 1 is a flowchart of an indoor positioning navigation method in an intelligent warehouse management system according to embodiment 1 of the present invention;

FIG. 2 is a zone division diagram of a warehouse to which the navigation method of FIG. 1 is applied;

FIG. 3 is a schematic plan view of region (v) in FIG. 2;

FIG. 4 is a schematic diagram of an apparatus for acquiring volume information according to the navigation method of FIG. 1;

FIG. 5 is a schematic view of the ultrasonic sensor B of the apparatus of FIG. 4 detecting cargo;

FIG. 6 is a schematic view of the ultrasonic sensor B of FIG. 5 detecting the angle of the cargo;

FIG. 7 is a front view of the device of FIG. 4;

FIG. 8 is a flow chart of the operation of the apparatus of FIG. 4;

FIG. 9 is a diagram of a mathematical model for solving the positioning of the intelligent vehicle in region (v) of the navigation method of FIG. 1;

FIG. 10 is a schematic view of a shelf of the warehouse of FIG. 2;

FIG. 11 is a flow chart of a portion of the navigation method of FIG. 1 for planning a driving path of an intelligent vehicle;

FIG. 12 is a flow chart of another portion of the navigation method of FIG. 1 for planning a driving path of the intelligent vehicle;

fig. 13 is a system function diagram of an indoor positioning navigation model in the intelligent warehouse management system according to embodiment 1 of the present invention.

Detailed Description

In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

Example 1

Referring to fig. 1,2 and 3, the present invention provides an indoor positioning navigation method in an intelligent warehouse management system, which is used for navigating an intelligent vehicle 7 for accessing goods 5 in a warehouse, and includes the following steps (step S1-step S19), and may further include step S20 and step S21. In this embodiment, the warehouse is provided with a plurality of rectangular areas, each rectangular area is provided with a plurality of roads arranged in a net shape, and two adjacent roads intersect at one node. Meanwhile, the number of the intelligent vehicles 7 is multiple, so that each intelligent vehicle 7 is defined as one ant, and the intelligent vehicles 7 form one ant colony.

And step S1, acquiring the volume information of the goods 5, the position information of the intelligent vehicle 7 and the vacancy information of the warehouse.

In this embodiment, the method for acquiring the volume information of the cargo 5 includes the following steps (step S101-step S104).

Step S101, measuring the length L of the cargo 5 at the inlet: firstly, carrying out ultrasonic ranging on the goods 5 through an ultrasonic transmitting end in the length direction of the goods 5 and starting timing; judging whether the ultrasonic ranging difference of two adjacent ultrasonic transmitting ends is greater than a threshold value I; thirdly, when the ultrasonic ranging difference is larger than the threshold value one, ending timing and obtaining scanning time; calculating the product of the moving speed of the goods 5 in the length direction and the scanning timeObtaining the length L of the cargo 5; i.e. L ═ v_TT，v_TT is the moving speed of the goods 5 and the scanning time.

Referring to fig. 4, specifically, the first ultrasonic sensor 1 (ultrasonic sensor a) is fixedly disposed on the detecting frame 2 for detecting the shape of the cargo 5, and is in a constant detecting state, and it is determined whether the distance difference between two adjacent distance ranges is smaller than the threshold m 1. If the distance measurement difference is smaller than the threshold m1, the boundary FH of the cargo 5 is not considered to be scanned, that is, the cargo 5 is not conveyed to the object form detection rack; if the distance difference is greater than the threshold m1, it is considered that the boundary FH of the cargo 5 is scanned, i.e., the cargo 5 arrives at the detection rack.

And when the goods 5 arrive at the detection frame, starting a timer for counting. And (4) ranging is always carried out in the counting process, and whether the ranging difference between two adjacent times is smaller than a threshold value m1 is judged. If the distance measurement difference is smaller than the threshold value m1, the boundary EG of the cargo 5 is not scanned, namely the cargo 5 does not leave the detection frame, and the timer continues to count; if the distance measurement difference is greater than the threshold m1, the boundary EG of the cargo 5 is considered to have been scanned, i.e., the cargo 5 has left the shelf, the current value of the timer is read, the scanning time T is obtained, and the timer count is initialized. The speed of the intelligent vehicle 6 is constant to v_TThe length of the cargo 5 can be obtained.

Step S102, measuring the width W of the cargo 5 at the inlet: firstly, a second ultrasonic transmitting end is rotated to generate a first fan-shaped detection surface in the width direction of the goods 5 so as to perform ultrasonic distance measurement on the goods 5; judging whether the second ultrasonic ranging difference between two adjacent ultrasonic ranging of the ultrasonic transmitting end is larger than a second threshold value; thirdly, when the second ultrasonic ranging difference is larger than the second threshold value, defining the second ultrasonic transmitting end to scan the boundary of the goods 5; calculating the time difference delta t between two opposite boundaries of the cargo 5 scanned by the second ultrasonic transmitting end and the distance S between the second ultrasonic transmitting end and the two boundaries_αAnd S_β(ii) a Fourthly, calculating the included angle theta of the first fan-shaped detection surface_L(ii) a Wherein, theta_LV is the rotation angular velocity of the second ultrasonic wave transmitting end; calculating the width W of the goods 5; wherein the content of the first and second substances,

referring to fig. 5, specifically, when the second ultrasonic sensor 3 (the ultrasonic sensor B) detects that the cargo 5 arrives, the rotation speeds of the steering engine for scanning by starting the second ultrasonic sensor 3 and the third ultrasonic sensor 4 affect the time and the precision for obtaining the size of the object by scanning. In general, the smaller the rotation speed of the steering engine is, the longer the time for scanning and obtaining the size is, and the higher the precision is; the larger the rotation speed of the steering engine, the shorter the time for scanning to acquire the size, but the relatively speaking accuracy is affected. Therefore, the time and the precision requirement are comprehensively considered, the rotating speed of the steering engine is assumed to be v, the steering engine only needs to rotate 180 degrees (the size of one surface of a cube is scanned in the range of 180 degrees) in practical use, and the steering engine rotates 180 degrees to an initial position after rotating 180 degrees in one direction. The object boundary is continuously scanned in the rotation process, which is equivalent to twice scanning, and the average value of the two calculated object side lengths can be used as the length of one side of the current scanning.

When the two 3 ultrasonic sensors combine the horizontal rotary table controlled by the steering engine to rotate, the two 3 ultrasonic sensors are always in a working state, the round-trip time t when the obstacles are encountered is detected in real time, and the distance S between the ultrasonic module and the obstacles can be calculated according to the following formula assuming that the speed is fixed and does not become c:

taking the control of the second ultrasonic sensor 3 as an example, a scanning effect map in actual use is shown in fig. 5. The ultrasonic sensor II 3 rotates to the og direction from the oa direction in the counterclockwise direction under the action of the vertical rotating shaft B controlled by the steering engine, and rotates 180 degrees together. After the system is initialized, when the ultrasonic sensor A detects that the goods 5 come, the ultrasonic sensor A starts to rotate, a timer is started to count the rotating time t, and the rotated angle theta can be calculated according to the time under the uniform rotation; in the rotating process, the ultrasonic module always transmits and receives ultrasonic signals, and the distance s between the ultrasonic module and the obstacle is calculated in real time.

Initial positionThe distance meter for measuring the oa direction is S₁The data of the next measurement is counted as S₂And by analogy, calculating the difference between the distance measured at this time and the distance measured at the last time. In the process from the oa direction to the oc direction, the measured distance difference between two adjacent times is smaller than a threshold value m (set according to practical conditions, related to the placement position of the intelligent vehicle 7 in the warehouse and the size of the warehouse), and the boundary AD of the goods 5 is not considered to be scanned. Let the alpha-th calculated distance be S_αThe distance calculated at the alpha-1 st time is S_α-1The alpha-2 calculated distance is S_α-2If the relationship shown below exists:

the alpha-1 th time does not reach the boundary AD, the alpha-1 th time reaches the boundary AD, and the timer time t at the moment is recorded_αAnd round trip time t_α1，t_α1The distance S at this time can be calculated by substituting the above equation_αAgain, from the relationship between the angle of rotation θ and the speed of rotation v and time t: θ ═ vt

The angle theta of rotation in the alpha-th scan to the boundary is known_α＝vt_α。

Similarly, in the process of turning to the oc direction and the oe direction, the adjacent distance difference every two times is always smaller than the threshold m, and the BC boundary of the ABCD surface is considered not to be scanned. Suppose that the distance calculated for the beta-th time is S_βThe calculated distance at the beta-1 st time is S_β-1The beta-2 calculated distance is counted as S_β-2If there is a relationship shown by the following formula:

the result shows that the boundary BC is not scanned for the beta-1 th time, the boundary BC is scanned for the beta-th time, and the timer time t at the moment is recorded_βAnd round trip time t_β1，t_β1After the substitution, the distance S at the moment can be calculated_βFrom the relationship between the rotation angle θ and the rotation speed v and time t, it can be seen that the boundary is reached in the beta-th scanAngle of rotation theta_β＝vt_βThe path swept by the sensor module over the ABCD surface is effectively X-Y in fig. 5.

From the above analysis, four parameters θ can be derived_α、θ_β、S_α、S_βThe relation, t, is as follows_α1And t_β1Respectively, the round trip time, t, when scanning to two boundaries_αAnd t_βRespectively for the time of rotating to two borders, c is the propagation velocity of ultrasonic wave, and v is the rotational speed of steering wheel, can have:

converts the practical problem into known four parameters theta_α、θ_β、S_α、S_βThe problem of finding a side length W of a triangle is shown in the mathematical model diagram of fig. 6. Knowing the length S of two sides of the triangle_α、S_βAnd the size of the included angle theta_LThe side length W of the side opposite to the included angle can be obtained by using the cosine law, and the side length is the width of the cargo 5, as shown in the following formula:

step S103, measuring the height H of the cargo 5 at the inlet: firstly, a fan-shaped detection surface II is generated in the height direction of the goods 5 by rotating an ultrasonic transmitting end III so as to carry out ultrasonic distance measurement on the goods 5; judging whether the ultrasonic ranging difference three of three adjacent two times of the ultrasonic transmitting end is greater than a threshold value three; thirdly, defining the boundary of the goods 5 scanned by the third ultrasonic transmitting end when the third ultrasonic transmitting end is larger than the third threshold; calculating the distance S between the third ultrasonic transmitting end and the boundary of the goods 5_ψThe vertical distance S between the third ultrasonic transmitting end and the goods 5_ζ(ii) a Calculating the height H of the goods 5:

wherein H₂Is the height of the ultrasonic transmitting end from three distances H to the ground₁Is the height of the cargo 5 from the ground.

Referring to fig. 7, specifically, the height information may be obtained by scanning the right side surface of the object by using the third ultrasonic sensor 4 (ultrasonic sensor C) in cooperation with the vertical rotating shaft, and actually, only the size of the upper half portion (ranging from the horizontal position to 90 °) needs to be scanned, the lower half portion may be equivalent to the height difference between the specific installation position of the third ultrasonic sensor 4 and the intelligent vehicle 6, and the height H of the cargo 5 may be obtained by adding the heights of the two portions, and the scanning effect graph is shown in fig. 7.

The sensor C is activated for scanning only when the ultrasonic sensor a detects the arrival of the cargo 5. Wherein S_ζThe round trip distance from the transmission to the reception of the ultrasonic sensor three 4 at the initial position, S_ψTo scan the round trip distance to the boundary location, t_ζRound trip time of initial position, t_ψIs the round trip time when the boundary BC is reached. H_hAnd S_ψAnd S_ζThe method is characterized in that a right-angle triangle is formed together, the actual problem is converted into the hypotenuse and a right-angle side of the known right-angle triangle, and the other right-angle side can be obtained by utilizing the pythagorean theorem, wherein the relationship is as follows:

step S104, calculating the volume V of the cargo 5: v — lxwxh. Therefore, the length, width and height dimension information of the goods 5 on the intelligent vehicle 7 is obtained through sensor scanning, and the volume of the goods 5 is assumed to be V and the unit is cm³Then, the following relationship is shown: l × W × H (cm)³)

The V calculated by the relational expression is the volume of the goods 5, so that the function of conveying the goods to container layers with different sizes according to different volumes is conveniently realized. In addition, in the present embodiment, the operation of the ultrasonic sensor A, B, C is as shown in fig. 8.

Secondly, in this embodiment, the method for collecting the position information of the smart car 7 includes the following steps: step S105, two positioning modules are respectively arranged at two ends of a diagonal line of the rectangular area, and the positioning modules are in wireless communication with the intelligent vehicle 7; step S106, calculating the distance S between the intelligent vehicle 7 and the two positioning modules according to the communication time of the intelligent vehicle 7 and the two positioning modules₁And S₂(ii) a Step S107, calculating the coordinates of two pre-marked points according to the coordinates of the two positioning modules; wherein, the distance between each pre-marking point and the two positioning modules is S₁And S₂(ii) a Step S108, respectively judging whether the two pre-marked points are positioned on the advancing path of the intelligent vehicle 7; when one of the pre-marked points is located on the traveling path and the other pre-marked point is not located on the traveling path, determining that the pre-marked point on the traveling path is the measuring position of the intelligent vehicle 7; and when the two pre-marked points are both positioned on the travelling path, selecting the pre-marked point which does not appear on the travelling path at the previous time as the measurement position of the intelligent vehicle 7 according to the previous measurement position of the intelligent vehicle 7.

The warehousing area is divided by considering real-time monitoring of the position of the intelligent vehicle 7 in a complex warehousing environment. The method for acquiring the position information of the smart car 7 is described by taking one of the regions (c) as an example, and a schematic diagram of the storage environment of the region (c) is shown in fig. 3. The circles represent the major road nodes and the shelves are placed in a straight line position in sequence.

A 4463 wireless module, namely X in fig. 9, is placed at both end points of a diagonal line of the storage region (c)₁(x₁,y₁) And X₂(x₂,y₂) And the two 4463 wireless modules wake up and send signals to the intelligent vehicle 7 after receiving the signals sent by the intelligent vehicle 7. After the 4463 wireless transmission module on the intelligent vehicle 7 receives the signals, the time of the signals sent by each 4463 positioning tag reaching the intelligent vehicle 7 is recorded respectively. Namely X₁Time of arrival t of the emitted signal₁，X₂Time of arrival t of signal sent by positioning tag₂。

Constant velocity v of the signal transmitted by each 4463 location tag_xRespectively calculating the distance between the intelligent vehicle 7 and the two 4463 positioning labelsAnd (5) separating. Suppose that the smart car is 7 away from X₁Is a distance S₁7-X separation of intelligent vehicle₂Is a distance S₂Distance S between two tags₃The following relational expression is given when the coordinates of the smart car 7 are P (x, y).

The problem is converted into the known three-edge length and the coordinates of two points, and the position of the intelligent vehicle 7 can be calculated by adopting a two-point positioning method. The position coordinates of the intelligent vehicle 7 obtained by the two-point positioning method are not unique, and two solutions meeting the condition exist, such as P in FIG. 9₁(x₃,y₃) And P₂(x₄,y₄)。

For determining that the position of the smart car 7 is P in fig. 9₁Or P₂And the optimal path of the intelligent vehicle 7 needs to be combined for judgment. Suppose that the smart car 7 is traveling along the route of A → B → C → D in FIG. 9, since only P is present₁Point on the driving path of the intelligent vehicle 7, P₁The point is the current position of the intelligent vehicle 7; if P₁And P₂All on the driving path, the position is judged to be P by combining the measured position of the last intelligent vehicle 7₁Or P₂。

In this embodiment, the method for acquiring vacancy information of a warehouse includes the following steps: step S109, two wireless modules respectively corresponding to the two positioning modules are respectively arranged at two ends of a diagonal line of the rectangular area; step S110, when the intelligent vehicle 7 receives a goods taking and storing instruction, sending awakening information to the wireless module of each rectangular area so as to awaken the wireless module to acquire the position information of the intelligent vehicle 7 through the corresponding positioning module; and step S111, driving the wireless module to sleep when the intelligent vehicle 7 leaves the rectangular area.

The shelves are sequentially placed on a road (straight line in the figure) connected with the main nodes as shown in fig. 3, each shelf is divided into three layers, namely large goods 5, medium goods 5 and small goods 5 from bottom to top, and the schematic diagram of a single shelf is shown in fig. 10. In order to save cost, a wireless module is not arranged on each layer of each goods shelf to transmit vacancy information, the vacancy information of the goods layers of each goods shelf is recorded into a data processing subsystem (server) in advance, and when the intelligent vehicle 7 carries out the access operation of goods 5, all the vacancy information does not need to be uploaded, and only the specific vacancy information needs to be updated to the server in time.

Initially, a 4463 wireless module (with a known position) is respectively placed at opposite corners of the warehouse shown in fig. 9, the smart car 7 sends wake-up information to the wireless module in the area only after receiving a goods taking or stocking instruction, and wakes up the wireless module to collect position information and upload the position information (distance from two positioning tags) in time; the intelligent vehicle 7 sends a sleep instruction to the wireless module in the area after leaving the area, so that the wireless module can sleep to reduce power consumption. After the shape information of the goods 5 is scanned and acquired by the ultrasonic sensor group on the shape detection frame of the goods 5, the shape information is transmitted to the warehouse extension set by the wireless module mounted on the MCU, and then is uploaded to the server.

In the present embodiment, the smart car 7 has the number i (i ═ 1,2, 3.., 8), and the state s_i，s_iWhen 0 indicates the car is in idle state, s_iThe car is assigned a mission. For the same task, the idle trolley with the minimum number (the highest priority) is preferentially arranged to complete, and each time one trolley completes the task, the idle trolley is updated to be in an idle state, is added into a queue, and waits for the next task to be issued.

Step S2, judging whether the intelligent vehicle 7 in the idle state exists; wherein, the intelligent vehicle 7 in the idle state is defined as the idle intelligent vehicle 7.

If at least one idle intelligent vehicle 7 exists, step S3 is executed to determine whether goods 5 exist at the entrance of the warehouse and need to be stored.

If at least one target good one is stored at the warehouse inlet, step S4 is executed, a target area for the target good one and a plurality of target shelves are determined according to the vacancy information and the volume information, and an idle intelligent vehicle 7 is determined as a stock vehicle.

And step S5, planning the stock path of the stock vehicle from the warehouse entrance to the target area and the plurality of target shelves, and driving the stock vehicle to travel according to the stock path.

Step S6, it is determined whether the stock vehicle reaches the stock point.

When the inventory vehicle reaches the inventory point, step S7 is executed to store the plurality of target goods in the inventory point one by one.

Step S8, determine whether the goods 5 need to be taken out from the warehouse.

If the second target cargo needs to be taken out from the warehouse, step S9 is executed to determine whether the second target cargo is in the current area of the inventory vehicle.

And if the target cargo II is in the current area, the stock vehicle is used as a goods taking vehicle.

If the second target cargo is not in the current area, executing step S10 to determine whether there is an intelligent vehicle 7 in an idle state; if the intelligent vehicle 7 in the idle state exists, step S11 is executed, and according to the location information of the pickup point, it is determined that one idle intelligent vehicle 7 is the pickup vehicle.

Step S12, a goods taking path from the current position of the goods taking vehicle to the goods taking point is planned, and the goods taking vehicle is driven to travel to the goods taking point according to the goods taking path so as to obtain the target goods II.

In the present embodiment, step S20, which is executed after step S12, is also provided. And step S12, judging whether the goods taking vehicle reaches the goods taking point, if so, executing the next step (step S13), otherwise, executing step S21, and adjusting the running state of the goods taking vehicle according to the relation between the position of the goods taking vehicle and the goods taking path.

And step S13, planning a delivery path from the delivery point to a delivery port of the delivery vehicle, and driving the delivery vehicle to travel to the delivery port according to the delivery path.

And step S14, judging whether the goods taking vehicle arrives at the goods outlet.

When the goods taking vehicle reaches the goods outlet, step S15 is executed, the second target goods in the goods taking vehicle are taken out, and the goods taking vehicle is returned to the bin entrance.

If the goods 5 do not need to be taken out of the warehouse, step S16 is executed to return the stock vehicle to the inlet.

When the inventory vehicle does not reach the inventory point, step S18 is executed to adjust the running state of the inventory vehicle in real time according to the real-time position of the inventory vehicle and the inventory path.

In the existing warehouse management system, although the optimal path planning by adopting the genetic algorithm has the characteristics of simplicity, convenience, rapidness and strong fault tolerance, the following problems are urgently solved:

(1) the parameter optimization problem of the algorithm is that the performance of the algorithm is difficult to improve by adjusting parameters according to the actual operating environment; (2) the method is easy to fall into 'premature' convergence, and is difficult to avoid that the solution is concentrated near the excellent solution prematurely during searching, so that a better solution is difficult to find, and the accuracy of running according to the optimal path in actual transportation cannot be ensured; (3) the algorithm has low efficiency, needs to be executed for a long time to achieve convergence, has high complexity, and is not suitable for the timeliness requirement of positioning and navigation in the storage room; (4) the fusion problem of the genetic algorithm and other optimization algorithms is difficult to improve the algorithm performance by being fused with other algorithms.

When the simulated annealing algorithm is adopted for path planning, although the experimental performance has the characteristics of high quality, strong initial robustness and universality, the method is easy to implement, but has the following defects:

(1) the contradiction between the optimization effect and the calculation time exists, theoretically, the probability 1 can be guaranteed to be converged to the global optimal solution as long as the calculation time is long enough, but in actual use, due to the limitation of the calculation speed and the calculation time, the calculation result is difficult to guarantee to be global optimal, and the optimization effect is not ideal; (2) whether the equilibrium state is reached is difficult to judge at each temperature, and the number of Metropolis processes is difficult to control; (3) two annealing modes in the simulated annealing algorithm are adopted, T is not corrected according to the change of the given rule all the time, and the performance of the algorithm cannot be improved by adjusting parameters according to the actual running condition.

Although the traditional Dijkstra algorithm can find the point-to-point optimal path in a plurality of paths and can store the shortest path from a source node to any node in the paths in solving the point-to-point optimal path problem, the method has certain limitations. The method is only suitable for solving the problem of point-to-point shortest paths and is not suitable for solving the problem of traversing shortest paths of a plurality of nodes, in addition, for traversing search, the source nodes need to be traversed once again when being frequently switched, the shortest paths between the two nodes are sequentially obtained, but the shortest total path can not be ensured during traversing, and the algorithm complexity is higher.

Based on the above problems, the present embodiment employs a highly efficient, highly accurate and less complex path planning method for planning the stocking path or the picking path to calculate the optimal travel path from the departure point to the target shelf for navigating the smart car 7. The embodiment is designed to adopt an improved ant colony algorithm to carry out multi-point access path planning so as to calculate the optimal driving path from the departure place to the destination. The ant colony algorithm has the following advantages: the ant colony algorithm is an algorithm combining distributed computation, a positive feedback mechanism and greedy search, and has strong capability of searching better solutions. Positive feedback can quickly find a better solution, the probability of 'premature' convergence is reduced to a great extent by distributed computation, and greedy search is helpful for finding out an acceptable solution in the early stage in the search process; secondly, the ant colony algorithm has strong parallelism, each ant simultaneously searches for a path, and the solving efficiency is high; the system has good expandability, and can cooperate through pheromone instead of direct communication among individuals, so that the system communication overhead increased along with the increase of individuals in the system is very small. Compared with simulated annealing algorithm and genetic algorithm, the ant colony algorithm has the highest quality of the found solution and requires fewer iterations to reach convergence, and has faster convergence speed, but the algorithm also has the following disadvantages: firstly, because the difference of pheromones on each edge is not obvious in the early stage, although the evolution can be carried out towards the optimal path through information exchange, when the population scale is large, a good path is difficult to find out from a large number of disorderly paths in a short time; the information positive feedback regulation mechanism enables the information quantity on a shorter path to be gradually increased, the information quantity on a better path can be obviously higher than the information quantity on other paths after a longer period of time, and the difference is more and more obvious along with the progress of the process, so that the better path is finally converged. This process generally takes a long time; the existence of pheromone volatilization factors rho (0< rho <1) can ensure that the information quantity on the path which is not searched is gradually less, thereby gradually reducing the searched probability to be near zero, reducing the global searching capability and easily missing the optimal solution; the probability that the searched path is selected again is increased due to overlarge rho value, the randomness of the algorithm and the global searching capability are influenced, the 'early-maturing' is caused, and the precision requirement of positioning and navigation is influenced; the convergence speed is reduced due to too small rho value, and the timeliness requirement of positioning navigation is influenced.

In view of the above two disadvantages, the present embodiment is designed to solve the following problems:

(1) in order to improve the solving efficiency, the realization complexity of the algorithm is reduced.

Combining Genetic Algorithm (GA) with Ant Colony Algorithm (ACA). The genetic algorithm has the capability of fast global search, but the feedback information in the system is not utilized, so that redundant iteration is often caused, and the solving efficiency is low; although the ant colony algorithm can enable pheromones of a shorter path to be obviously higher than those of other paths through a positive feedback mechanism of the pheromones, the early pheromones are deficient, and the algorithm speed is slow.

Considering advantage complementation, the random search, global convergence and rapidity of the genetic algorithm are used to generate initial pheromone distribution related to the problem. Then, the parallelism, the positive feedback mechanism and the high efficiency characteristic of the ant colony algorithm are fully utilized to carry out solving. The fused algorithm is superior to the ant colony algorithm in time efficiency and superior to the genetic algorithm in solving efficiency, and a heuristic algorithm with both time efficiency and solving efficiency is formed.

And secondly, considering the track limitation of the actual storage environment, the server limits the candidate set, and only adds a plurality of trunk nodes which can be directly reached by the intelligent vehicle 7 (with connection attributes in the figure 2) into the candidate set, so that the improvement of the candidate set further improves the solving efficiency.

(2) The method aims to solve the problems that partial optimality is trapped in the actual search of the ant colony algorithm too early and the search stagnation phenomenon occurs. The ant colony algorithm is combined with a mechanism capable of effectively avoiding premature convergence, and the ant colony algorithm with optimal performance is obtained to improve the precision.

One avoids search stalls, considering starting with influencing the probability used to select the next solution, which depends directly on the pheromone trajectory and heuristic information. The heuristic information is determined by the problem and cannot be changed, but the influence of the pheromone tracks can be limited, and the overlarge difference between the pheromone tracks in the algorithm operation process is avoided. Therefore, a mechanism for limiting the maximum value and the minimum value of the pheromone track in a maximum-minimum ant system (MMAS) is introduced, so that the pheromone tracks tau of all sides are limited_ij(t) with τ_min＜τ_ij(t)＜τ_max. The judgment conditions for achieving convergence are as follows: at each selection point, the trace amount on one of the solution elements is tau_maxThe trace on all other alternative solution elements is τ_min。

Introducing worst ant pheromone global updating rule in the best-worst ant system. The optimal solution is enhanced to a greater extent, and the worst solution is weakened, so that the pheromone difference between the edge belonging to the optimal path and the edge belonging to the worst path is increased, and the search is more concentrated near the optimal solution. If (r, s) is an edge on the worst ant path and is not an edge in the optimal ant path, the pheromone quantity is adjusted according to the following formula:

introducing self-adaptive regulation mechanism of pheromone volatilization factor rho. The optimal solution is reserved after each cycle, when the obtained optimal solution is not obviously improved in N cycles, the rho adjustment is reduced by 0.1, and the minimum value is set as rho_minPreventing too little decrease in convergence speed, i.e.:

in the embodiment, the distribution route with the minimum energy consumption is solved by taking the limited weight of the distribution vehicle as a main target and taking the minimum route energy consumption as a main target, and comprehensively considering the distribution time, wherein the distribution route with the minimum energy consumption and short time is obtained under the condition of considering the emergency degree of the goods. In the embodiment, the ant colony algorithm and the genetic algorithm are considered to be fused, the initial pheromone accumulation is rapidly generated by the genetic algorithm, the fitness function and the transition probability can be adaptively adjusted according to different distribution requirements, and then the optimal distribution path meeting the different distribution requirements is solved by utilizing the positive feedback search characteristic of the ant colony algorithm. Referring to fig. 11 and 12, the method for planning the inventory path or the pickup path includes the following steps (step S501-step S516).

Step S501, initializing a population, a code and a genetic variable, and determining a fitness function. In this embodiment, the cross probability pc, the mutation probability pm, and the maximum evolutionary algebra G are initialized_maxMinimum evolution algebra G_minMinimum rate of evolution G_ratioEvolution end algebra G_end. In this embodiment, the population size is set to N, and an initial population G is obtained, such that G_min＜G＜G_maxCoding according to actual problems, determining a fitness function f(s), and calculating the fitness value f of individuals in the population_i. When encoding, each distribution node is represented by one byte, such as 01 and 03. If the total weight of the cargo 5 is M, each distribution is carried outThe total weight of the goods 5 required by the node i is m_iThe distance from the current node to the next distribution node i is d_iThe delivery cost is denoted as s. The following relationship is given:

defining fitness

Then it is necessary to ask:

so that the fitness f is maximum

Step S502, the individuals in the parent group are sorted according to the fitness, so that the probability of selecting the individual with higher fitness is larger than the probability of selecting the individual with lower fitness. In this step, the selection operation is completed, so that the probability of selecting the individual with higher fitness in the parent group is higher, the probability of selecting the individual with lower fitness is lower, and the better individual can be selected with higher probability.

Step S503, judging whether mutation operation needs to be carried out on the ant colony; when the ant colony needs to be subjected to mutation operation, executing step S504, performing reverse mutation on the ant colony, and selecting new individuals and parent individuals according to the fitness;

when the ant colony does not need to be mutated or the step S504 is completed, step S505 is executed to determine whether the ant colony needs to be crossed. The process of the crossover operation is as follows:

selecting two father strings, and randomly selecting a mating area as follows:

old1＝12|3456|789

old2＝98|7654|321

② add region of old2 before old1, add region of old1 before old 2:

old1^＝7654|123456789

old2^＝3456|987654321

and thirdly, deleting the same numbers in old1^ and old2^ as the mating area in sequence to obtain the final two sub-strings:

new1＝765412389

new2＝345698721

wherein, mutation operation: reverse mutation methods, such as 1-2-3-4-5-6, break between intervals 2-3 and 5-6, and insert in reverse order to become 1-2-5-4-3-6.

When the ant colony needs to be subjected to the crossover operation, executing step S506, performing the crossover operation on the ant colony, and selecting the new individual and the parent individual according to the fitness; and when the ant colony does not need to be subjected to cross operation or the step S506 is completed, executing the step S507, and judging whether the ant colony meets a preset fusion condition. In order to make the population evolve towards a direction with higher fitness, the crossed and mutated individuals better than the parents are reserved into a new parent population, and the individuals with the fitness value lower than that of the parent band are eliminated and are not put into the parent band population. In addition, the embodiment compares the new individuals with the individuals in the original parent population, performs the good and bad replacement of the individuals according to the result, and selects the individuals with high fitness as the new children of the next generation.

When the ant colony satisfies the fusion condition, step S508 is performed to convert the solution obtained by the genetic algorithm into an initial pheromone distribution value on the path of the ant colony and define a path pheromone range. Initializing an initial value of pheromone on a path in the ant colony algorithm by using a better solution generated by a genetic algorithm, and limiting the concentration of the pheromone on the path according to the following formula in order to prevent the search from getting early due to the fact that the early difference of the pheromone on the path is large and a better solution cannot be found:

when the ant colony does not satisfy the fusion condition, step S502 is performed. If is G_min＜G＜G_maxAnd the rate of evolution G_end＞G_ratioThen, the process goes to step S502, (i.e., the step ofThe evolution rate of successive generations is higher than a set termination threshold), otherwise proceed to the next step.

Step S509, initializing the number of nodes, the number of ants, the cycle number and the pheromone volatilization factor that the ant colony needs to traverse, randomly placing the ants on each node, and emptying a taboo table of the ant colony;

step S510, judging whether the loop iteration times of the ant colony population are not less than the maximum loop times;

when the number of loop iterations of the population is not less than the number of loop iterations, executing step S511, and outputting a search result;

when the loop iteration number of the population is smaller than the loop number, step S512 is executed, and each ant selects the next node according to the following state movement rule formula:

namely:

wherein, tau_ij(t) is the pheromone track, table_kAs the tabu table, allowed_kIs a candidate set; lambda [ alpha ]_jM is the degree of urgency of the cargo 5_jTotal weight of cargo 5 required for node j, d_ijIs the distance from the current node to the next node. The tabu table has the function of preventing repeated access to the node, and the strategy of the candidate set is to limit the selectable range which can be used as the next mobile target node, avoid a large number of redundant iterations and reduce the complexity of the algorithm. In the conventional ant colony algorithm, only the distance factor is considered, and the total transportation time can only be guaranteed to be the shortest in the actual cargo 5 storing and taking process (assuming that the delivery trucks transport at a constant speed, and the time difference for storing and taking the cargo 5 at each cargo 5 access point is not considered).

The embodiment is improved in two aspects:

the distribution route and the distribution energy consumption are comprehensively considered, the cargo demand weight of each distribution node is also considered, under the condition that the distribution route is short, the intelligent vehicle 7 can preferentially store heavier cargos 5 during multi-point inventory, and the intelligent vehicle 7 can preferentially take lighter cargos 5 during multi-point pick-up, so that the energy consumption is reduced.

Secondly, the nodes of the goods 5 needing to be accessed urgently are weighted according to the emergency degree (priority) of the goods 5, and the emergency goods 5 are accessed preferentially while the energy consumption is considered.

The fitness function and the state transition probability can be adaptively adjusted according to the following three practical conditions, and the intelligent vehicle 7 carries out multipoint inventory:

1) when the intelligent vehicle 7 is used for stock or taking goods, the following steps are provided:

wherein f is the fitness of the fitness function;

2) when the inventory energy consumption of the intelligent vehicle 7 is the lowest, the following steps are carried out:

when 7 energy consumptions of getting goods of intelligent car are minimum, have:

3) when the intelligent vehicle 7 needs to take the cost, the time and the emergency of the goods 5 into consideration, the following steps are provided:

when 7 intelligent vehicle gets goods and need consider expense, time of use and goods 5 emergency, have:

λ_ithe larger the cargo 5, the more urgent is the cargo 5 λ for ordinary cargo _i1, may be λ depending on the urgency of storage of the cargo 5_i＝2,3,...。

In step S513, after the kth ant traverses a node of one circle and returns to its starting point, the pheromone on the path traveled by the kth ant is locally updated. In this embodiment, the pheromone range is also updated, and the limiting formula is:

τ (r, s) ← ρ · τ (r, s) + (1- ρ) · Δ τ (r, s) (equation 4)

And repeating the steps S512 and S513 until all ants traverse the nodes of one week to return to the starting point.

Step S514, according to the maximum fitness f_maxAnd minimum f_minAnd updating the optimal path length of the iteration and the pheromone of the road section included in the optimal path length, and the worst path of the iteration and the pheromone of the road section included in the worst path. Similarly, in step S514, the present embodiment further limits the pheromone range; wherein the content of the first and second substances,

τ(r,s)←τ(r,s)+·Δτ(r,s)

in the formula, L_gbThe optimal path length for this iteration;

where (r, s) is an edge that belongs to the worst path and not to the optimal path, L_worstFor the worst path length of this path finding, L_bestThe optimal path length for the path finding is the attenuation coefficient of the pheromone on the worst path and the edge which does not belong to the optimal path after the circulation is finished, and the value range is [0, 1 ]]. The parameter is a parameter that represents the attenuation coefficient of the pheromone on the edge of the worst path selected and not belonging to the optimal path after the end of each cycle. The larger the value of the coefficient is, the larger the attenuation of the pheromone to the edge only belonging to the worst path in the cycle is; the smaller the value, the smaller the attenuation degree of pheromone on the edge only belonging to the worst path in the cycle.

Step S515, reset the position of the mth ant as the starting point, and set the empty tabu table_kJudging whether the value of the information volatilization factor needs to be adjusted or not;

when the value of the information volatilization factor needs to be adjusted (if no more optimal path is found in 3 consecutive iterations, the value of the information volatilization factor is adjusted), step S516 is executed, and step S510 is executed after the value of the information volatilization factor ρ is adjusted according to the following formula:

To sum up, the indoor positioning navigation method in the intelligent warehouse management system of the embodiment has the following advantages:

according to the navigation method, the volume information of the goods 5 is collected, the storage area and the goods shelves of the goods 5 are determined according to the volume of the goods 5, a plurality of vacant goods shelves are selected as target goods shelves according to the volume of the goods 5 and the vacant information during storage, a plurality of goods taking points can be selected as the target goods shelves during goods taking, the shortest conveying path is determined, the intelligent vehicle 7 is guided to achieve multi-point access operation of an indoor storage environment by matching two-point positioning, so that the storage utilization rate is improved, meanwhile, the idle intelligent vehicle 7 close to the goods 5 is selected to be accessed by collecting the position information of the intelligent vehicle 7 and the vacant information of a warehouse, the access time of the intelligent vehicle 7 can be shortened, the storage efficiency is improved, and the intelligent vehicle 7 in the warehouse is convenient to navigate.

In the present embodiment, when planning an inventory route or a pickup route, an initial pheromone distribution related to a problem is first generated using a random search, global convergence, and rapidity of a genetic algorithm. Then, the parallelism, the positive feedback mechanism and the high efficiency characteristic of the ant colony algorithm are fully utilized to carry out solving. And finally, fusion is carried out, the fused algorithm is superior to the ant colony algorithm in time efficiency and superior to the genetic algorithm in solving efficiency, and a heuristic algorithm with both time efficiency and solving efficiency is formed. In addition, in the embodiment, the track limitation of the actual storage environment is considered, the candidate set is limited, only a plurality of trunk nodes which can be directly reached by the intelligent vehicle 7 are added into the candidate set, and the solution efficiency is further improved through the improvement of the candidate set. Meanwhile, in the embodiment, the ant colony algorithm is combined with a mechanism capable of effectively avoiding premature convergence, the ant colony algorithm capable of obtaining the optimal performance is used for improving the precision, the search stagnation is avoided, the optimal solution is enhanced to a greater extent, the worst solution is weakened, the pheromone difference between the side belonging to the optimal path and the side belonging to the worst path is increased, the search is concentrated near the optimal solution, in addition, an adaptive adjustment mechanism of pheromone volatilization factors is introduced, the minimum path energy consumption is taken as a main target, the distribution time is comprehensively considered, and the distribution path with the minimum energy consumption and short time under the condition of considering the emergency degree of goods is solved.

The acquisition of the size of the goods 5 in the existing warehousing management system mainly adopts an image processing method, acquired image information is transmitted to a processor for processing, the information amount is large, the cost is high, and the size of the goods 5 cannot be dynamically detected, in the embodiment, form information for conveying the goods 5 and position information of the intelligent vehicle 7 are determined, the size of the goods 5 is dynamically scanned through an ultrasonic module, and the position information of the intelligent vehicle 7 is acquired through two-point positioning.

Example 2

Referring to fig. 13, the present embodiment provides an indoor positioning navigation model in an intelligent warehouse management system, which is used for navigating an intelligent vehicle 7 for storing and taking goods 5 in a warehouse, and includes a data acquisition subsystem, a data transmission subsystem, and a data processing subsystem.

The data acquisition subsystem is used for acquiring the volume information of the goods 5, the position information of the intelligent vehicle 7 and the vacancy information of the warehouse. In this embodiment, the data acquisition subsystem includes an intelligent vehicle 7 position information acquisition module and a cargo 5 form information acquisition module. The cargo 5 form information acquisition module comprises an ultrasonic microphone A, an ultrasonic microphone B, an ultrasonic microphone C, STM32 single chip microcomputer, a cargo 5 form detection frame and a 4463 wireless module, wherein the STM32 single chip microcomputer receives and analyzes length, width and height information detected by the ultrasonic microphone A, B, C and generates volume information. The goods 5 form detection frame is used for the goods 5 to pass through, and is provided with an intelligent vehicle 6 for the goods 5 to pass through. The data acquisition subsystem of this embodiment may adopt the method for acquiring the volume information of the cargo 5 in embodiment 1 to acquire data.

The data transmission subsystem is used for updating the position information, the volume information and the vacancy information in real time according to the information acquired by the data acquisition subsystem and the established low-power-consumption transmission mechanism of the two-point positioning model. The data transmission subsystem of the embodiment combines the low power consumption transmission advantage of the Si4463 module, obtains the real-time coordinates of the intelligent vehicle 7 by a two-point positioning method, and provides a measurement and calculation basis for whether the intelligent vehicle 7 deviates from the shortest access path. The specific data transmission method may adopt the method described in embodiment 1, and the warehouse extension receives and uploads the volume information and the distance information.

The data processing subsystem is used for the contents of step S2 through step S21 in embodiment 1, and a plurality of discrete units may be provided to perform each step. The data processing subsystem selects a plurality of vacant shelves as target shelves according to the volume of the goods 5 and the vacancy information during inventory, screens a plurality of goods taking points as the target shelves during goods taking, determines the shortest conveying path, and is matched with two-point positioning to guide the intelligent vehicle 7 to realize multi-point access operation of the indoor storage environment. The extension set transmits the signal to the server, and the server screens out a plurality of target shelves and an optimal distribution path after analyzing the information and navigates the intelligent vehicle 7 to carry out the access operation of the multi-point goods 5. Aiming at the path planning problem of multipoint access, the data processing subsystem is combined with the advantages of the ant colony algorithm, accelerates the accumulation of initial pheromones through the introduction of the genetic algorithm, promotes the rapid global convergence of the ant colony algorithm, and obtains the shortest access path.

Example 3

The present embodiments provide a computer terminal comprising a memory, a processor, and a computer program stored on the memory and executable on the processor. The processor implements the steps of the indoor positioning navigation method in the intelligent warehouse management system of embodiment 1 when executing the program. When the method in embodiment 1 is applied, the method can be applied in a software form, for example, a program designed to run independently is installed on a computer terminal, and the computer terminal can be a computer, a smart phone, a control system, other internet of things equipment, and the like. The method of embodiment 1 may also be designed as an embedded running program, and installed on a computer terminal, such as a single chip microcomputer.

Example 4

The present embodiment provides a computer-readable storage medium having a computer program stored thereon. When the program is executed by the processor, the steps of the indoor positioning navigation method in the intelligent warehouse management system of embodiment 1 are implemented. When the method of embodiment 1 is applied, the method may be applied in the form of software, such as a program designed to be independently run by a computer-readable storage medium, which may be a usb disk designed as a usb shield, and the usb disk is designed to be a program for starting the whole method through external triggering.

The above description is only for the purpose of illustrating the preferred embodiments of the present invention and is not to be construed as limiting the invention, and any modifications, equivalents and improvements made within the spirit and principle of the present invention are intended to be included within the scope of the present invention.

Claims

1. An indoor positioning navigation method in an intelligent warehouse management system, which is used for navigating an intelligent vehicle for storing and taking goods in a warehouse, is characterized by comprising the following steps:

a step S5 of planning an inventory path of the inventory vehicle from the inlet to the target area and the target shelf, and driving the inventory vehicle to travel according to the inventory path;

step S6, determining whether the stock vehicle reaches a stock point;

executing step S7 when the inventory vehicle arrives at the inventory point, storing the target item one at the inventory point;

step S8, determining whether goods need to be taken out of the warehouse;

if the second target cargo needs to be taken out from the warehouse, executing step S9 to determine whether the second target cargo is in the current area of the stock vehicle;

step S14, judging whether the goods taking vehicle arrives at the goods outlet;

when the goods taking vehicle reaches the goods outlet, executing step S15, taking out the target goods II in the goods taking vehicle, and returning the goods taking vehicle to the bin inlet;

step S17, updating the status of the stock vehicle or the delivery vehicle at the inlet to an idle status;

the warehouse is provided with a plurality of rectangular areas, a plurality of roads which are arranged in a net shape are arranged in each rectangular area, and two adjacent roads are intersected at a node; the number of the intelligent vehicles is multiple, each intelligent vehicle is defined as one ant, and the intelligent vehicles form an ant colony; the planning method of the stock path or the goods taking path comprises the following steps:

step S501, initializing a population, a code and a genetic variable, and determining a fitness function;

step S502, sequencing the individuals in the parent group according to the fitness, so that the probability of selecting the individual with higher fitness is greater than the probability of selecting the individual with lower fitness;

step S503, judging whether mutation operation needs to be carried out on the ant colony;

when the ant colony needs to be subjected to mutation operation, executing step S504, performing reverse mutation on the ant colony, and selecting new individuals and parent individuals according to the fitness;

when the ant colony does not need to be subjected to mutation operation or the step S504 is completed, executing the step S505, and determining whether the ant colony needs to be subjected to crossover operation;

when the ant colony needs to be subjected to the crossover operation, executing step S506, performing the crossover operation on the ant colony, and selecting the new individual and the parent individual according to the fitness;

when the ant colony does not need to be subjected to cross operation or the step S506 is completed, executing the step S507, and judging whether the ant colony meets a preset fusion condition;

when the ant colony meets the fusion condition, executing step S508, converting a solution obtained through a genetic algorithm into an initial pheromone distribution value on a path of the ant colony, and defining a path pheromone range; when the ant colony does not meet the fusion condition, executing step S502;

step S510, judging whether the cycle iteration times of the ant colony are not less than the maximum cycle times;

when the number of loop iterations of the population is not less than the maximum number of loop iterations, executing step S511, and outputting a search result;

wherein, tau_ij(t) is the pheromone track, table_kAs the tabu table, allowed_kIs a candidate set; lambda [ alpha ]_jM is the degree of urgency of the cargo_jTotal weight of cargo required for node j, d_ijThe distance from the current node to the next node;

step S513, after the kth ant traverses a circle of nodes and returns to the departure point, the pheromone on the path of the kth ant is locally updated;

repeating the step S512 and the step S513 until all ants traverse a circle of nodes to return to the starting point;

step S514, according to the maximum fitness f_maxAnd minimum f_minUpdating the optimal path length of the iteration and the pheromones of the road sections included in the optimal path length, and the worst path of the iteration and the pheromones of the road sections included in the worst path of the iteration;

when the value of the information volatilization factor needs to be adjusted, step S516 is executed, and after the value of the information volatilization factor ρ is adjusted according to the following formula, step S510 is executed:

when the value of the information volatilization factor does not need to be adjusted, directly executing the step S510;

wherein, when the intelligent vehicle is used for stock or taking goods the shortest time, have:

wherein f is the fitness of the fitness function;

2. the method for indoor location navigation in the smart warehouse management system according to claim 1, wherein in step S508, the path pheromone range is defined by the formula:

in step S513, the pheromone range is also limited, and the limiting formula is:

in step S514, the pheromone range is also limited; wherein the content of the first and second substances,

τ(r,s)←τ(r,s)+·Δτ(r,s)

in the formula, L_gbThe optimal path length for this iteration;

3. The method for indoor location and navigation in the intelligent warehouse management system according to claim 1, wherein the method for indoor location and navigation further comprises the steps of:

4. The indoor positioning and navigation method in the intelligent warehouse management system according to claim 1, wherein the method for acquiring the volume information of the goods comprises the following steps:

step S101, measuring the length L of goods positioned at the warehouse inlet;

in the length direction of the goods, performing ultrasonic ranging on the goods through an ultrasonic transmitting end and starting timing;

judging whether the ultrasonic ranging difference of two adjacent ultrasonic transmitting ends is greater than a first threshold value;

when the ultrasonic ranging difference is larger than the threshold value one, ending timing and obtaining scanning time;

calculating the product of the moving speed of the goods in the length direction and the scanning time to obtain the length L of the goods;

step S102, measuring the width W of goods at the bin entrance;

a first sector detection surface is generated in the width direction of the goods by rotating a second ultrasonic transmitting end so as to perform ultrasonic ranging on the goods;

judging whether the second ultrasonic ranging difference between two adjacent ultrasonic transmitting ends is larger than a second threshold value;

when the second ultrasonic ranging difference is larger than the second threshold, defining the second ultrasonic transmitting end to scan the boundary of the goods; calculating the time difference delta t between the two opposite boundaries scanned by the second ultrasonic transmitting end and the goods and the distance S between the second ultrasonic transmitting end and the two boundaries_αAnd S_β；

Calculating the included angle theta of the first fan-shaped detection surface_L(ii) a Wherein, theta_LV is the rotation angular velocity of the second ultrasonic wave transmitting end;

calculating the width W of the cargo; wherein the content of the first and second substances,

step S103, measuring the height H of the goods at the warehouse inlet;

by rotating an ultrasonic transmitting end III, a fan-shaped detection surface II is generated in the height direction of the goods so as to perform ultrasonic ranging on the goods;

judging whether the ultrasonic ranging difference III of three adjacent ultrasonic ranging of the ultrasonic transmitting end is larger than a threshold value III;

when the third ultrasonic transmitting end is larger than the third threshold, defining the boundary of the cargo scanned by the third ultrasonic transmitting end; calculating the distance S between the third ultrasonic transmitting end and the boundary of the goods_ψThe vertical distance S between the third ultrasonic transmitting end and the goods_ζ；

Calculating the height H of the cargo:

step S104, calculating the volume V of the cargo: v — lxwxh.

5. The indoor positioning and navigation method in the intelligent warehousing management system according to claim 1, wherein the warehouse is provided with a plurality of rectangular areas, each rectangular area is provided with a plurality of roads arranged in a net shape, and two adjacent roads intersect at a node; the method for acquiring the position information of the intelligent vehicle comprises the following steps:

s105, respectively placing two positioning modules at two ends of a diagonal line of the rectangular area, wherein the positioning modules are in wireless communication with the intelligent vehicle;

step S106, calculating the distance S between the intelligent vehicle and the two positioning modules according to the communication time of the intelligent vehicle and the two positioning modules₁And S₂；

Step S107, calculating the coordinates of two pre-marked points according to the coordinates of the two positioning modules; wherein, the distance between each pre-marking point and the two positioning modules is S₁And S₂；

Step S108, respectively judging whether the two pre-marked points are positioned on the advancing path of the intelligent vehicle;

when one of the pre-marked points is located on the traveling path and the other pre-marked point is not located on the traveling path, judging that the pre-marked point on the traveling path is the measuring position of the intelligent vehicle;

and when the two pre-marked points are positioned on the advancing path, selecting the pre-marked point which does not appear on the advancing path in the previous time as the measuring position of the intelligent vehicle according to the measuring position of the intelligent vehicle in the previous time.

6. The method for indoor location navigation in an intelligent warehouse management system according to claim 5, wherein the method for acquiring vacancy information of the warehouse comprises the following steps:

step S109, two wireless modules respectively corresponding to the two positioning modules are respectively arranged at two ends of a diagonal line of the rectangular area;

step S110, when the intelligent vehicle receives a goods taking and storing instruction, sending awakening information to the wireless module of each rectangular area so as to awaken the wireless module to acquire the position information of the intelligent vehicle through the corresponding positioning module;

and step S111, driving the wireless module to sleep when the intelligent vehicle leaves the rectangular area.

7. An indoor positioning navigation model in an intelligent warehouse management system, which is used for navigating an intelligent vehicle for storing and taking goods in a warehouse, comprises a data acquisition subsystem, a data transmission subsystem and a data processing subsystem, and is characterized in that,

the data acquisition subsystem is used for acquiring the volume information of the goods, the position information of the intelligent vehicle and the vacancy information of the warehouse;

the data transmission subsystem is used for updating the position information, the volume information and the vacancy information in real time according to the information acquired by the data acquisition subsystem;

the data processing subsystem is configured to perform the steps of:

step S6, determining whether the stock vehicle reaches a stock point;

executing step S7 when the inventory vehicle arrives at the inventory point, storing the plurality of target goods in the inventory point one by one;

step S8, determining whether goods need to be taken out of the warehouse;

step S12, planning a goods taking path from the current position of the goods taking vehicle to the plurality of goods taking points, and driving the goods taking vehicle to travel to the goods taking points according to the goods taking path so as to obtain the target goods II;

step S13, planning a delivery path from the last delivery point to the delivery port of the delivery vehicle, and driving the delivery vehicle to move to the delivery port according to the delivery path;

step S14, judging whether the goods taking vehicle arrives at the goods outlet;

wherein f is the fitness of the fitness function;