Disclosure of Invention
The present invention is directed to overcoming the above-mentioned drawbacks and deficiencies of the prior art and providing an integrated automated device for visual recognition, automatic sorting, pick-and-place, voice prompt, voice query, etc. of storage multi-mode integrated information (color, shape, contour, number, letter, text, two-dimensional code), which is connected to a management system.
Realized the pronunciation interaction, voice command, pronunciation inquiry information, through information acquisition device, gather operation personnel information, through visual device discernment extranal packing, the integrated information (colour, shape, profile, digital code, trade mark pattern, letter, characters, two-dimensional code) of article, digital code and two-dimensional code on the scanning goods, the automatic medicine preparation of confirming with the goods information of vision collection is double-accurate to the inquiry management goods, offers, reduces numerous and diverse degree, the complexity of staff's work.
The invention provides a cargo scanning device, a positioning device and a pusher, which pass through a rail type sliding device.
The invention provides a goods operation table device which is used for identifying storage, external packaging and subpackaging goods through machine vision.
The invention provides a goods conveyor belt device, which controls the direction of a delivery box (a flowing medicine box) on the conveyor belt according to path planning, automatically combines and stores goods on each task position.
The invention provides a voice recognition method for voice interaction, voice instruction, voice recognition, voice synthesis and the like.
The invention solves the problems and adopts the technical scheme that:
automatic change storing device includes:
the main control system module is used for realizing the main control of the machine, the interaction between the voice module and a user, the voice module, the visual identification module, the information acquisition device, the operation platform device, the cutting and cutting device, the storage (medicine) pushing sliding device, the scanning device, the conveying belt device and the article taking-out device.
And the voice module is used for text-to-voice, voice-to-text, voice guidance and man-machine voice interaction.
The visual identification module is connected with the camera device and the code scanner and used for collecting the medicine box image and identifying the trademark, the color and the characters.
And the information acquisition and reading device is used for scanning and reading the information of the user card.
The storage (medicine) pushing sliding device is used for storing the medicine, the medicine storage pusher and the sliding device along the rail.
The operation platform device is used for cutting, clipping and subpackaging storage medicines.
And the cutting and cutting device is connected with the operation table device and is used for cutting and cutting the articles (medicines).
And the scanning device is used for scanning two-dimensional codes, digital codes, RF tags and the like attached to articles (medicine boxes, books and warehouse goods).
And the article taking-out device is used for taking out and distributing articles from the storage box.
The conveyer belt device is used for conveying articles and the storage box.
Voice devices such as a voice acquisition device and a microphone are connected with the main system to establish a user-machine voice interaction library, and a voice recognition method, voice synthesis, character voice conversion, voice recognition, voice query, voice management of stored objects, medicines and the like are utilized.
The system comprises an information acquisition device, two-dimensional codes and digital code scanning and reading devices, wherein the information acquisition device is in communication connection with a main system, and the scanning devices are arranged below all sliding rail openings and above operation platforms to read the two-dimensional codes and the digital codes of cargoes. The information acquisition device and the reading device are arranged at the taking-out port of the delivery box and used for scanning the user information card and receiving the stored articles and the medicines from the taking-out port.
The storage (medicine) management system and the user management system are connected with the camera, the scanning reader and the information acquisition and reading device, and management personnel (users) and other various information such as storage (medicine) two-dimensional codes, digital codes, personnel biological information, RFID information and the like are intelligently identified through an improved machine learning method and an improved neural network method.
And the vision device is arranged above the position of the operation platform corresponding to each sliding track port. The device is used for collecting images of stored articles (medicine boxes, books, warehouse goods and the like) and identifying trademarks, colors and characters. The visual recognition module comprises: the storage (medicine) recognition, the shape recognition, the quantity recognition, the character recognition and the trademark image recognition are realized, compared with the information scanned and read by the code scanner, the code scanner reads the storage information and confirms the character and trademark on the storage box in a double mode, the neural network method is improved, and the storage (medical medicine and the like) is recognized and managed.
Storing and thrust unit with slip track, its storing promotes slide device and includes: the device comprises a storage device, a pusher, a sliding rail crossing device and a four-way article overturning device arranged on the lower side of the sliding rail crossing device. Storing promotes slider for the storage of article promotes target position article, pushes the slip track according to the quantity of storing unit and arranges the roll-over table in, and four-way storing turning device of slip track crossing device downside can be according to four side rotations about the positive and negative, scans storing (medicine) two-dimensional code, and the digital code utilizes vision device to gather storing (medicine box) image and discernment trade mark, colour, characters.
The operating platform device for independently subpackaging and cutting to finish various operations is used for supporting the operations on the operating platform device. The method comprises the following steps: thrust unit, cutting device, storing medicine box device.
And the pushing device is used for pushing the pusher on the operation table to push the storage (medicine).
The cutting device adopts crosscut, rip cutting, synthesizes the cutting, cuts into slices, tailors the device for the storing, and the medicine is according to the antithetical couplet and the cutting of linking vertically, and discernment storing (tablet etc.) quantity partial shipment is stored in operation platform storing box, waits to convey and takes the shipment case (mobile medical kit) that flows to reach the storing box and correspond the position, pops out storing (medical kit) to the mobile shipment case (medical kit) of target.
The automatic delivery device comprises an automatic article delivery device and a delivery box (mobile medicine box), wherein a conveyor belt is arranged on the lower layer of an operation platform and used for conveying the delivery box (mobile medicine box). The conveying device comprises a conveying device, a goods outlet box and a direction control device. And the delivery box (the mobile medicine box) reaches a target object taking position of the task according to the planned path, the code scanning reads the information of the target storage box to be matched with a single target or a plurality of targets in the delivery box (the mobile medicine box) task, and the storage box is popped up to the delivery box (the mobile medicine box).
A method for visually identifying a stored substance (medicine) and the quantity thereof comprises the following steps:
s1, creating a mathematical model according to the characteristic items of the storage (medicine) including color, outline, shape, attached numerical code, characters and special identification.
And S2, extracting color, outline and digital code words, and identifying the image characteristics corresponding to the image by the special identification words, wherein the special identification words comprise color, number, outline, digital code, words and characters.
And S3, converting the characteristics of color, number, letter, character, special trademark identification value and the like into input data.
And S4, establishing a mathematical model of the image characteristics, and inputting the characteristic values of the detection items.
And S5, improving the weight optimizer, and quickly training the image to obtain an output value.
And S6, identifying the storage (medicine) target accurately according to the output color, number, letter, character and special identification, and returning the target positioning target to the target position (storage medicine placement position).
A dynamic optimized shipment (flowing medicine box) path planning method adopts an optimized method to plan shipment (medicine dispensing) tasks, and aims to finish all tasks in the shortest time. The optimization method comprises the following steps: a genetic calculation method and an improvement method thereof, a tabu search calculation method and an improvement method thereof, a simulated annealing calculation method and an improvement method thereof, an ant colony calculation method and an improvement method thereof, a calculation method for particle beam optimization and an improvement method thereof, a neural network calculation method and an improvement method thereof, an evolution method and an improvement method thereof, and one or more combination methods thereof. The method comprises the following specific steps:
the mathematical model is as follows:
job jobmFor each jobmComprising n task tasksn。
The push of task n slides to the operation table, and the operation time of cutting operation
job
mIn-process task
NMove to task
LTime of
Slave tasknThe time waiting for the completion of the preceding job is waitingInter taskn waitingtime。
The time required to complete a single job is calculated as follows:
an objective function:
Max{job1 total,job2 total,job3 total...jobm total}
further, the improved ant colony method dynamically configures operation to realize optimal shipment task planning, and aims to complete all tasks in the shortest time, and comprises the following steps:
s1, initially unpacking a plurality of jobs, each job being composed of tasks contained therein, job jobsm addDynamically adding into the planned task group.
Job joba without additionm addThe task movement path corresponding to the original solution is as follows, jobam-solution={job1,job2,job3,...jobm}
jobm-solution={{M1,M5,M3,M6}{M2,M5,M7,M6},{M3,M5,M4,M8},...{M3,M5,M4,M9}}。
S2, calculating adding operation time point, and re-planning task with incomplete pathnThe location area M.
S3, calculating M different in current time pointmNumber of tasks M not yet completedaccount m1/M inverse ratio of number of tasks per regionaccount m,
Defining informationElement Mm effort=1/Maccount m。
That is, the number of tasks is small, pheromones are inversely proportional to the number of tasks, and pheromones are large.
S4, calculating and setting an evaporation speed parameter task of pheromonespeed nThe number of tasks increases and the pheromone decreases.
Mm effort=1/Maccount m-taskspeed n*n
The number of tasks decreases and pheromones increase.
Mm effort=1/Maccount m+taskspeed n*n
S5, calculating M for each M aream effortAccording to Mm effortFrom multiple to few orders, jobam addThe task at least selects a moving path according to the pheromone of the M area, calculates and finds the jobsm addReplanning the path to obtain a new candidate solution jobm-new。
S6, generating a new candidate solution jobam-new。
S7 Jobm-new<jobm-old,
Will jobmNew enters the evolution iteration as a new candidate solution.
And S8, outputting the optimal solution according to the iteration times.
The management system comprises a storage management module, a storage positioning module, a storage identification module, user management and a user reminding module.
Drawings
Fig. 1 is a schematic view of a machine module in the present application, with reference to fig. 1:
100-machine master control system module; 200-a workbench module; 300-image acquisition visual identification module;
400-information acquisition scanning module; 500-cutting the cropper module; 600-medicine, store, push slide;
700-a voice module; 800-conveyor means and a mobile medicine chest;
FIG. 2 is a block diagram of the machine structure of the present application, a U-shaped line drawing, referenced 2;
201-a medicine storage box, a storage box; 202-positioning the pushing device;
203-workbench (pushing device, cutting device, storage box and medicine box device);
204-a vision device; 205-a scanning device;
206-conveyor means; 207-delivery box (mobile case);
208-medicine outlet (cargo outlet); 209-machine master control system; 210-a directional control device;
211-speech means; 212-a sliding track;
fig. 3 is an X-shaped line drawing in the present application, with reference to fig. 3:
201-a medicine storage box, a storage box; 202-positioning the pushing device;
203-workbench (pushing device, cutting device, storage box and medicine box device);
204-a vision device; 205-a scanning device;
206-conveyor means; 207-delivery box (mobile case);
208-medicine outlet (cargo outlet); 209-machine master control system; 210-a directional control device;
211-speech means; 212-a sliding track;
fig. 4 shows a work table working apparatus according to the present application, and reference numeral 4:
203-workbench (pushing device, cutting device, storage box and medicine box device)
204-a vision device; 205-a scanning device; 207-delivery box (mobile case);
Detailed Description
A machine device system and method, the device comprising: the main control system of the machine, the voice device module, the camera image acquisition module, the information acquisition module, the medicine storage box, the positioning pushing device, the cutting device, the storage medicine box device, the vision device, the conveyer belt device, the flowing medicine box, the medicine outlet, the direction control device and the machine main control system.
The technical solution implemented in the present application is to solve the above technical problems as follows
This scheme is mainly through the storage tank, and location thrust unit realizes the storage of article, and the operation is tailor in automatic sorting, shipment.
Through parameter setting of the directional sound recognition device and the microphone module, human-machine voice interaction is achieved and goods and medicine information is inquired through voice by means of voice recognition, voice-character conversion, voice awakening and the like.
The scheme mainly identifies the medicine boxes and trademarks by setting the camera and image parameters and utilizing an improved machine learning method and a neural network method, and identifies the quantity of the medicine boxes and the trademarks by extracting articles, the colors, the shapes and the outlines of the tablets.
The machine utilizes information acquisition reading device, inquires goods, medicine information, improves the efficiency of management, has improved the flexibility of inquiry greatly.
The delivery box (the medicine box flows) is realized through the conveyor belt device, the delivery box (the flowing medicine box) and the direction control device, the direction can be controlled to reach the delivery position of the operation platform, and the merging, sorting and delivery of the goods at each task position are completed.
This scheme is mainly through location thrust unit in the storage tank, location medicine storage position, promotes storing (medicine) and goes out the medicine via the sliding tray.
The robot is used for replacing a human to complete complex sorting and delivery operations, the efficiency is improved, and the labor cost is saved. The pressure of personnel operation is greatly reduced, and the working efficiency is improved.
In order to better understand the technical solutions, the present invention will be further described in detail with reference to the following examples and drawings, but the embodiments of the present invention are not limited thereto.
Example 1;
as shown in fig. 1 and fig. 2, an embodiment of an artificial intelligence machine for pharmacies, warehouses and libraries is implemented by the method comprising the following steps:
the main control system 700, the main control system 100 module is used for realizing the main control of the machine, the voice module 700 and the user interact, the visual recognition module 300, the information acquisition device 400, the operation platform device 200, the cutting and cutting device 500, the storage (medicine) pushing sliding device 600, the scanning device 500, the conveyor belt device 800 and the article taking-out device.
The voice module 700 is used for text-to-speech, speech-to-text, speech guidance, and man-machine voice interaction.
The visual recognition module 300 is connected with the camera device 300 and the scanning device 205, collects medicine box images and recognizes trademark color characters.
The information acquisition and reading device 400 (doctor card ) scans the doctor card, doctor card and patient information.
The drug storage pushes the sliding device 600, and the storage (drug) slides to the device along the rail under the push of the pusher.
The operation table device 200 is used for storing (medicine) split charging, cutting and cutting.
And a cutting and clipping device 500 connected to the operation table device for cutting and clipping the articles (medicines).
And the code scanner 400 is used for scanning the code of the medicine attached to the medicine box.
A storage (medicine) taking-out port 208, a storage (medicine) taking-out device and a storage (medicine) dispensing device, which are used for taking out and dispensing the medicine from the storage box.
And a conveyor device 800 for conveying the medicine boxes.
Example 2:
the main control system 209 of the robot is connected with the camera 204 and the information acquisition and reading device 205 to acquire medicine and storage information. And reading the information of the manager. The main control system 209 communicates with the voice device 211, and the machine and the user interact with each other by voice, so as to inquire the medicine information and the voice medical orders.
The machine main control system 209 is in communication connection with the camera image acquisition module 204, and acquires a medicine box storage outer package image, storage (medicine) information and a medicine storage package image.
According to the task of getting the stored articles (medicines) issued by the machine main control system device 209 to each storage (medicine) getting place, the medicine storage pushing 202 sliding device 212 slides along the rail 212 to the operation platform device 203 area, and the vision device 204 and the scanning device 205 are utilized to collect the images of the stored medicine boxes and identify the color and characters of the trademarks. Storage (medicine) information. Utilize scanning device 205 to scan the medicine box and read storing (medicine) information, realize two accurate medicine discernments. The quantity of discernment medicine, according to doctor's advice and storing (medicine) distribution information, the cutting ware module carries out the medicine and divides the dress, cuts.
The flowable medicine tank 207 waits for the target articles (medicines) to be put into and out of the container (flowable medicine tank) 207 from the work table to each medicine taking position according to the task plan. All the delivery target tasks are completed, and the stored articles (medicines) are taken out from the stored article (medicine) take-out port 208 and distributed.
Example 3:
on the basis of embodiment 1, the machine main control system module 209 is connected with the vision module 204 and the scanning device 205, and identifies the drug storage and the quantity thereof, and the double-precision confirmation method of the visually identified drug storage information and the drug storage information read by scanning the code comprises the following steps:
as shown in fig. 2 and 3, the mathematical model is created according to the characteristic items of the drug storage, including color, contour, shape, attached digital code, attached two-dimensional code, characters and special identification. And extracting the image characteristics corresponding to the color, the contour, the digital code, the two-dimensional code and the character, wherein the special identification comprises the color, the number, the contour, the digital code, the two-dimensional code, the character and the special identification character identification image. The features are converted into input data. And improving a weight optimizer, and quickly training the image to obtain an output value. The medicine box information of accurate discernment medicine storing target, the medicine name, the trade mark, the target location target returns and snatchs the target position (storing medicine place the position). Utilize scanning device scanning medicine box extranal packing to read storing (medicine) information, realize visual identification storing (medicine) information and sweep the two accurate affirmations of storing (medicine) information that the code scanning read, realize storing (medicine) discernment.