Intelligent sorting simulation teaching system and method

Abstract

The application relates to the technical field of artificial intelligence application, and provides an intelligent sorting simulation teaching system, which comprises a sorting pad, a sorting table and a feedback host; the sorting pad is divided into a sorting area and a plurality of storage areas; the sorting area is used for placing the sorting table; the sorting table is used for placing a target sample to be sorted; and the storage areas are used for placing the target sample after sorting; the sorting table is provided with a data collector; each storage area is provided with a sensor; the data collector and each sensor send collected data and sensing data to the feedback host; the feedback host confirms target sample characteristics of the target sample based on the collected data, determines a target storage area where the target sample is located based on the sensing data, compares the target sample characteristics and the target storage area with a pre-established storage relationship, determines whether a sorting result of the target sample meets an expectation, and outputs a feedback result.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of artificial intelligence application technology, in particular to an intelligent sorting simulation teaching system and method. BACKGROUND

[0002] With the rapid development of computer application technology, artificial intelligence technology has been applied to daily life from theoretical research, such as intelligent customer service, face recognition, license plate recognition, etc.

[0003] Based on the visibility of artificial intelligence technology application and the continuous maturity of related simulation development technology, the demand for popular science education is also expanding, and various artificial intelligence teaching and training projects have emerged.

[0004] However, in the existing artificial intelligence teaching project, attention is paid to the interest and operability of the teaching project, that is, the implementation of a simple program control mechanical process is used to show the implementation process of basic artificial intelligence. In the existing experiment process, the student and the teacher can only observe the result of the program running with their eyes, and the teaching feedback is not clear and lacks guidance. SUMMARY

[0005] In order to meet the teaching project in keeping the interest of learning, and also can be clear for the student teaching feedback and guidance, the present application embodiment provides an intelligent sorting simulation teaching system and method.

[0006] On the one hand, the intelligent sorting simulation teaching system provided by the present application embodiment comprises a sorting pad, a sorting table and a feedback host;

[0007] The sorting pad is divided into a sorting area and a plurality of storage areas, the sorting area is used for placing the sorting table, the sorting table is used for placing the target sample to be sorted, and the storage area is used for placing the target sample after sorting; the sorting table is provided with a data collector, each storage area comprises a hard layer and a flexible layer, the hard layer and the flexible layer are connected at the edge to form a hollow area, a sensor is arranged in the hollow area, the sensor is arranged on the flexible layer, and the data collector and each sensor are signal connected with the feedback host and send collection data and sensing data to the feedback host based on the signal connection;

[0008] The feedback host confirms the target sample characteristics of the target sample based on the collection data, determines the target storage area where the target sample is located according to the sensing data, and compares the target sample characteristics and the target storage area with a pre-established storage relationship to determine whether the sorting result of the target sample meets the expectation, and outputs a feedback result.

[0009] Based on the sensors built in the sorting pad and the data collector on the sorting table provided by the above system, real sensor data and collected data corresponding to the target sample can be collected. Thus, the feedback host can analyze the operation process of the mechanical arm based on the data to determine whether the operation of the mechanical arm meets the expectation and output the feedback result in real time, so that the student can intuitively obtain real-time experimental feedback through the feedback host without relying on naked eye observation. Thus, accurate and clear teaching feedback is realized, and the student can be guided to pay attention to the program running through the feedback result.

[0010] Further, by arranging the sensors on the flexible layer, the stability of the sensors can be ensured, thereby reducing the influence of the sensors on the sensing data and improving the data accuracy.

[0011] In an implementation, the system further comprises a mechanical arm and a mechanical arm controller. The mechanical arm is connected with the mechanical arm controller and fixed in an operation area of the sorting pad, and the operation area is adjacent to the sorting area. The mechanical arm controller is programmed with a mechanical arm control program for controlling the mechanical arm to transfer the target sample from the sorting table to the corresponding storage area, wherein the control program determines the storage area corresponding to the target sample based on the storage relationship.

[0012] By planning the installation positions of the mechanical arm and the mechanical arm controller based on the sorting pad, the position planning of each component of the system can be realized, and the inaccurate data collection caused by the placement of the component positions can be reduced. The mechanical arm transfers the target sample based on the control program, and the feedback host analyzes the operation result of the mechanical arm based on the storage relationship set in the control program, thereby ensuring the accuracy of the analysis result.

[0013] In an implementation, the system further comprises a sample identifier. The sample identifier is programmed with an identification program based on an artificial intelligence algorithm. The sample identifier identifies the characteristics of the target sample based on the identification program and sends the identification result to the mechanical arm controller. The mechanical arm controller determines the storage area corresponding to the target sample based on the identification result and the storage relationship.

[0014] The sample identifier identifies the characteristics of the target sample and sends the identification result to the mechanical arm controller, so that the mechanical arm controller can sort and transfer the target sample based on the identification result. Thus, the simulation process of intelligent sorting can be realized.

[0015] In an implementation, the sample identifier is connected with the feedback host and sends the identification result to the feedback host. The feedback host compares the identification result with the characteristics of the target sample and generates and feeds back the accuracy rate of the identification program based on the comparison result.

[0016] Therefore, after receiving the identification result sent by the sample identifier, the feedback host can also analyze the accuracy of the identification result based on the collection data. In this way, the feedback host can not only determine whether the operation of the mechanical arm meets the expectation, but also can feed back the identification accuracy of the identification program, so that the key link of the entire sorting experiment process has a corresponding feedback result output, providing full-process feedback for students.

[0017] In an embodiment, the sorting pad further has an induction controller built-in, which is connected with each sensor through a signal line and receives the sensing data sent by each sensor. After filtering each sensing data, the sensing data is sent to the feedback host.

[0018] By filtering the initial data obtained by each sensor in advance through the induction controller, the interference of invalid data on the feedback host can be reduced, and the calculation burden of the feedback host can be reduced.

[0019] In an embodiment, the sorting table further comprises a base, the base comprises a first layer plate and a second layer plate arranged oppositely, the first layer plate is coincident with the center point of the second layer plate, and the area of the first layer plate is smaller than that of the second layer plate, the data collector is fixed to the first layer plate through a connecting structure, and the surface of the second layer plate away from the first layer plate is provided with an anti-skid structure.

[0020] By designing the sorting table into a circular table shape and setting the anti-skid structure, the installation stability of the sorting table can be ensured, and the influence of the shaking of the sorting table on the clarity of image collection and the operation of the mechanical arm during the sample sorting simulation process can be reduced, and the interference of external factors on the identification program and the control program can be eliminated as much as possible.

[0021] In one aspect, the intelligent sorting simulation teaching method provided by the embodiments of the present application is applied to a feedback host, and comprises the following steps:

[0022] receiving sensing data sent by a built-in sensor of a sorting pad and collection data sent by a data collector in real time;

[0023] confirming a target sample feature of a target sample based on the collection data;

[0024] determining a target storage area where the target sample is located according to the sensing data;

[0025] comparing the target sample feature and the target storage area with a pre-established storage relationship to determine whether a sorting result of the target sample meets an expectation, and outputting a feedback result.

[0026] In an embodiment, the real-time receiving of the sensing data sent by the built-in sensor of the sorting pad comprises the following steps:

[0027] Real-time receiving sensing data sent by the sensing controller built in the sorting pad, wherein the sensing data is collected by each sensor and obtained through filtering of the sensing controller.

[0028] In an implementation, the filtering method comprises:

[0029] Each sensor sends the collected initial sensing data to the sensing controller in real time;

[0030] The sensing controller respectively compares each initial sensing data sent by each sensor with previous data;

[0031] If the initial sensing data changes positively compared with the previous data, the initial sensing data is taken as the sensing data obtained through filtering.

[0032] In an implementation, the determining of the target storage area where the target sample is located according to the sensing data comprises:

[0033] Based on the difference between the collection data and the receiving time of the sensing data, judging whether the sensing data is valid;

[0034] If valid, determining the storage area where the sensor sending the sensing data is located as the target storage area.

[0035] In this implementation, the feedback host judges whether the sensing data is valid through the time difference, so as to finally obtain valid sensing data, which can eliminate the influence of human interference on the feedback result, ensures the accuracy of the feedback result, and improves the fault tolerance of the system.

[0036] In summary, the embodiments of the present application at least have the following beneficial technical effects:

[0037] 1. The feedback host analyzes the real data collected by the data collector and the sensor, the recognition result output by the recognition program, and the theoretical operation result of the control program, and outputs the corresponding feedback result, so that the key nodes in the whole sorting simulation process have clear feedback content output, and the students.

[0038] 2. By setting the sensor on the flexible layer, the stability of the sensor can be ensured, thereby reducing the influence of the sensor on the sensing data and improving the data accuracy.

[0039] 3. Each key link in the sorting simulation process is visible, which guides to improve the students' cognition of process monitoring. BRIEF DESCRIPTION OF DRAWINGS

[0040] Figure 1 The drawing illustrates the structure schematic diagram of the intelligent sorting simulation teaching system provided by the embodiments of the present application;

[0041] Figure 2 Fig. 1 shows a schematic diagram of a sorting table structure in an embodiment of the present application;

[0042] Figure 3 Fig. 2 shows a schematic diagram of a sorting mat structure in an embodiment of the present application;

[0043] Figure 4 Fig. 3 shows a schematic diagram of a sorting table structure in an embodiment of the present application; Figure 3 Fig. 4 shows a schematic diagram of a cross-section along the A-A' cross-section line;

[0044] Figure 5 Fig. 5 shows a flowchart of a smart sorting simulation teaching method provided in an embodiment of the present application. DETAILED DESCRIPTION

[0045] The embodiments of the present application will be described in detail below with reference to the drawings.

[0046] The embodiments of the present application disclose a smart sorting simulation teaching system and method, which can provide experimental feedback to students through a feedback host, so that the students or teachers can directly know the running condition of the sorting simulation teaching system according to the feedback information. The embodiments of the present application will be described below in combination with specific application examples.

[0047] Please refer to Figure 1 , the smart sorting simulation teaching system 100 includes a sorting mat 110, a sorting table 120, a feedback host 130, a mechanical arm 140, a mechanical arm controller 150, a sample identifier 160, and a plurality of samples 170, wherein the sorting table 120 is used to carry the samples 170 to be sorted, each sample 170 has different morphological characteristics, including but not limited to color, shape, marking, weight, etc., for simulating different morphological goods, wherein only one sample is shown in the figure, and other samples can be placed in a designated area before being sorted, and the students can manually place the samples to be sorted on the sorting table 120, or set the control program of the mechanical arm to transfer the samples to be sorted from the designated area to the sorting table 120 through the mechanical arm.

[0048] Please refer to Figure 2 , the sorting table 120 includes a base 121 and a data collector 122, the base 121 is in the shape of a circular table and is hollow, including a first layer plate 121a and a second layer plate 121b arranged oppositely, and a surrounding plate 121c connected between the first layer plate 121a and the second layer plate 121b, which form a hollow circular table shape, the center points of the first layer plate 121a and the second layer plate 121b coincide, and the area of the first layer plate 121a is smaller than that of the second layer plate 121b. The data collector 122 is arranged inside the sorting table 120 and is fixed to the surface of the first layer plate 121a close to the second layer plate 121b through a connecting structure, and the surface of the second layer plate 121b away from the first layer plate 121a is provided with an anti-skid structure.

[0049] In an example, the data collector 122 is a Radio Frequency Identification (RFID) reader, and each sample is provided with an RFID tag, and the RFID tag is pre-written with identification information for uniquely identifying the corresponding sample. Thus, when the sample is placed on the sorting table 120, the data collector 122 can read the identification information of the sample and send the collected data to the feedback host. Preferably, the data collector 122 can be selected as a low-frequency RFID reader to prevent mis-collection of samples not placed on the sorting table 120. Further, a signal isolation layer can be added to the surrounding plate 121c to prevent the radio frequency signal from being emitted from the side of the sorting table 120, thereby further preventing misidentification.

[0050] The data collector 122 can be mounted on the first layer plate 121a through a connecting structure, which can be a glue layer provided between the data collector 122 and the first layer plate 121a, or a buckle fixed on the first layer plate 121a. The data collector 122 is fixed on the first layer plate 121a through the buckle, and the data collector 122 can send the collected data to the feedback host 130 in a wireless manner.

[0051] The anti-slip structure provided on the second layer plate 121b can be composed of a plurality of anti-slip pads, which are uniformly and spacedly arranged on the surface of the second layer plate 121b to achieve stable placement of the sorting table 120 on the sorting mat 110. Optionally, a metal isolation sheet is embedded in the anti-slip pad, and a plurality of anti-slip pads form a signal isolation between the second layer plate 121b and the sorting mat 110 to prevent signal interference between the sorting mat 110 and the data collector 122.

[0052] Referring to Figure 3 The main body of the sorting mat 110 is a rubber mat with a certain thickness, which is used to carry other components in the intelligent sorting simulation teaching system 100 and define the placement positions of the other components. Specifically, the sorting mat 110 is divided into a sorting area 111, a plurality of storage areas 112, an operation area 113, an identification area 114, and the like, each area is distinguished by an identification line and marked with an area description. Each storage area 112 is provided with a corresponding number (such as C1, C2...C6 in the figure) to distinguish from each other. Figure 3 In an example, the sorting area 111, the operation area 113, and the identification area 114 are arranged along the same straight line, and the sorting area 111 is located between the operation area 113 and the identification area 114. The plurality of storage areas 112 are distributed on both sides of the sorting area 111, so that the layout of each area is compact, and the operation of the mechanical arm and the image acquisition of the sample identifier are facilitated.

[0053] The sorting area 111 is used to place the sorting table 120. The sorting area 111 can be provided with a clamping groove matched with the anti-skid pad to accommodate the anti-skid pad. The clamping groove can enhance the action feedback of the students during the process of building the simulation system, improve the interest, and further improve the stability of the sorting table 120.

[0054] The operation area 113 is used to place the mechanical arm 140 and the mechanical arm controller 150. The mechanical arm 140 is connected with the mechanical arm controller 150 and can be fixed in the operation area in a detachable manner. For example, the edge of the base of the mechanical arm 140 is provided with a suction cup. The suction cup is adsorbed on the sorting pad 110 to fix the mechanical arm 140. At the same time, the fixing mode different from the sorting table 120 can improve the practical ability and interest of the students. The mechanical arm controller 150 is programmed with a mechanical arm control program for controlling the mechanical arm 140 to transfer the sample on the sorting table 120 to the corresponding storage area 112. The control program is written by the students and burned into the mechanical arm controller 150. The control program sets the corresponding relationship between the sample features and the storage area. For example, if the sample feature is red, it corresponds to the storage area 112 numbered as C1.

[0055] The identification area 114 is used to define the installation position of the sample identifier 160. The sample identifier 160 runs an identification program based on an artificial intelligence algorithm. The identification program can be used to identify the features of the sample on the sorting table 120 and send the identification result to the mechanical arm controller. Based on the identification result and the storage relationship set in the control program, the mechanical arm controller determines the corresponding storage area 112 of the sample on the sorting table 120.

[0056] In an example, the sample identifier 160 can be an artificial intelligence development board, which at least includes a mainboard, and a camera unit and a display unit integrated on the mainboard. The mainboard can identify the sample image collected by the camera unit based on the pre-stored identification program to output the sample features. The mainboard can also control the display unit to display the identification result, so that the students can watch the content of the identification result, thereby directly obtaining the running situation of the identification program, to improve the visibility of the system in the identification stage. In order to enable the camera unit of the sample identifier 160 to accurately capture the sample image on the sorting table 120, the camera unit of the sample identifier 160 is further provided with a view angle simulator for displaying the shooting range of the camera unit. In an example, the view angle simulator can be implemented based on a visible light emitter to display the shooting range of the camera unit by emitting visible light. In this way, the students can know the shooting range of the camera unit according to the illumination range of the visible light, thereby adjusting the shooting range to ensure that the camera unit can completely capture the sample on the sorting table.

[0057] Reference Figure 4, each storage area 112 is internally provided with a sensor 112d, and Figure 4 In the example, the sensor is a stress sensor, and the storage area 112 includes a hard layer 112a and a flexible layer 112b arranged above and below, respectively. The hard layer 112a can be made of hard materials such as tempered glass, hard plastic, etc., and the flexible layer 112b is made of materials such as rubber and silicone that can easily deform. The hard layer 112a and the flexible layer 112b are connected at the four peripheral edges to form a closed hollow structure 112c, which is filled with inert gas. The stress sensor is arranged close to the flexible layer 112b to collect the stress changes of the flexible layer 112b. For example, when a sample is placed on the storage area 112, the weight of the sample acts on the upper hard layer 112a, and the deformation of the hard layer 112a is very small. Therefore, under the action of the sample weight, the inert gas will be compressed, causing the flexible layer 112b to deform. In this way, the stress sensor can collect the stress changes of the flexible layer 112b. By arranging the stress sensor close to the flexible layer 112b, compared to arranging it close to the hard layer, it is more stable, thus reducing the influence of the stress sensor itself on the collection results, and improving the accuracy of the sensing data.

[0058] In another example, the sensor can also be a weighing sensor that can collect the weight of the sample on the storage area 112.

[0059] The sorting table 120 is also internally provided with a sensing controller connected to each sensor through signal lines to receive the sensing data sent by each sensor. The sensing controller and the signal lines are internally provided in the rubber pad. The sensing controller can send the sensing data to the feedback host 130 in a wireless manner.

[0060] Based on the above system, the student can pre-upload the prepared identification program to the sample identifier 160, burn the control program to the robot arm controller 150, and after completing the installation of each component according to the indication on the sorting table 120, start the sorting simulation: first, randomly select a target sample from multiple samples and place it on the sorting table 120, then the sample identifier 160 performs feature recognition based on the collected target sample image and outputs the feature recognition result to the robot arm controller 150, so that the robot arm controller 150 controls the robot arm based on the recognition structure and the control program to realize the transfer of the target sample on the sorting table 120 to the corresponding storage area. At the same time, after reading the identification information of the target sample, the data collector 122 on the sorting table 120 can send it to the feedback host 130, and the corresponding sensor of the storage area 112 also sends the sensing data to the feedback host 130, so that the feedback host 130 can output the feedback result based on the identification information and the sensing data to indicate whether the identification of the target sample is accurate and whether the operation of the robot arm is as expected. Thus, real-time feedback of the actual operation process is realized.

[0061] Please refer to Figure 5 The intelligent sorting simulation teaching method provided by the embodiments of the present application will be described below from the perspective of the feedback host 130.

[0062] The method specifically comprises the following steps:

[0063] S510, real-time receiving of the sensing data sent by the sorting pad built-in sensor and the collection data sent by the data collector.

[0064] In implementation, the feedback host 130 real-time receives the sensing data sent by the sensing controller built-in the sorting pad, wherein the sensing data is collected by each sensor and obtained through filtering of the sensing controller.

[0065] Specifically, when the sensor detects that the sensing data changes, the sensor sends the sensing data to the sensing controller, that is, each sensor real-time sends the collected initial sensing data to the sensing controller, and the sensing controller determines the sensor number sending the initial data after receiving the initial sensing data, and compares the initial sensing data with the last data sent by the sensor; if the initial sensing data changes positively compared with the last data, the initial sensing data is taken as the filtered sensing data and sent to the feedback host 130. In an implementation, the positive change of the data refers to that the initial sensing data is greater than the last data, for example, a sample is put into the storage area.

[0066] It can be understood that since the students may be younger and have a stronger curiosity for various things, in the simulation teaching process, the students may move the samples on the sorting table 120 or the storage area 112 by themselves, at this time, the sensor and the data collector will normally send the related data to the feedback host 130, therefore, the feedback host 130 needs to verify the data after receiving the sensing data and the collection data, so as to ensure that the data is generated in the running process of the simulation teaching system, and not generated by the students' interference.

[0067] In an embodiment, after receiving the collection data, the feedback host 130 saves the collection data in a first queue and records the receiving time. After receiving the sensing data, the feedback host 130 saves the sensing data in a second queue and records the receiving time, and calculates the time difference between the receiving time of the sensing data and the latest recorded receiving time of the collection data in the first queue. If the time difference meets a preset condition, the sensing data is determined as valid data, i.e., data generated by the system running, wherein the preset condition includes that the time difference is consistent with the operation time length of the mechanical arm. The operation time length of the mechanical arm can be pre-collected, i.e., the time consumed from when the sample is placed on the sorting table 120 to when the sample is transferred to the storage area 112 in the normal running process of the system. It can be understood that, since the distances from the mechanical arm to different storage areas are different, the operation time length can take an interval value, and the two endpoints are the time consumed to the nearest storage area 112 and the time consumed to the farthest storage area 112, respectively. If the time difference does not meet the preset condition, the sensing data is directly discarded, i.e., deleted from the second queue. Meanwhile, if no valid sensing data is received within a preset time length, the collection data is marked as invalid, and the feedback host 130 displays the invalid data in the feedback result.

[0068] In another embodiment, the sensing data can be verified by setting different weights for different samples in combination with the collection data. Specifically, the corresponding sensing data of each sample is pre-collected, i.e., each sample is placed in a storage area, and the corresponding theoretical sensing data is recorded. By setting different weights for different samples, the corresponding theoretical sensing data of each sample is also different. When the feedback host 130 receives the collection data, the sample identification information is determined according to the collection data, and the corresponding theoretical sensing data of the sample is determined according to the sample identification information. The received sensing data is verified based on the theoretical sensing data. If they are the same, it is determined as valid sensing data. If no valid sensing data is received within a preset time length, the collection data is marked as invalid. In this way, the valid sensing data can be more accurately determined, and invalid calculation of the feedback host 130 due to invalid data can be avoided.

[0069] In this way, by verifying the collection data and the sensing data by the feedback host 130, the subsequent feedback calculation can be effectively avoided due to human interference, thereby improving the fault tolerance of the system.

[0070] S520, confirming the target sample feature of the target sample based on the collection data.

[0071] Specifically, the identification information of the target sample can be determined by collecting data, and all target sample features corresponding to the target sample, including shape, color, identification, etc., can be obtained by querying the data record. It is worth noting that when the students write the identification program, they can select one or more sample features as the identification dimension, and the output result of the identification program corresponds to the content corresponding to the selected identification dimension. For example, if the color and shape are selected as the identification dimension, the output result of the identification program can be red and square. Since the programs written by each student are different, the feedback host 130 can obtain all sample features corresponding to the target sample based on the collected data as the target sample features.

[0072] S530, determining the target storage area where the target sample is located according to the sensing data.

[0073] After the feedback host 130 verifies the valid sensing data based on the above method, the storage area where the sensor sending the sensing data is located can be determined as the target storage area.

[0074] S540, comparing the target sample features and the target storage area with the pre-established storage relationship to determine whether the sorting result of the target sample meets the expectation, and outputting the feedback result.

[0075] In an implementation, the storage relationship includes sample features and corresponding storage areas, which are defined by the students when writing the control program of the mechanical arm. The feedback host 130 can collect the storage relationship from the students before the system runs. For example, the feedback host 130 can collect the storage relationship from the students when the students complete the writing of the control program. The target sample features and the target storage area are real data obtained based on the data collector and the sensor, and the storage relationship is theoretical data. The target sample features can be used to query the storage relationship to determine the theoretical storage area corresponding to the target sample features, i.e., the storage area to which the mechanical arm should transfer the target sample. If the theoretical storage area is consistent with the target storage area, it indicates that the control program of the mechanical arm runs as expected, i.e., the sorting result meets the expectation. In this way, the feedback host 130 can feedback the result and record it.

[0076] In another implementation, the sample identifier 160 can send the identification result to the feedback host 130 after generating the identification result. The feedback host 130 can verify the identification result based on the target sample features to determine the accuracy of the identification program. If the target sample features include all the sample features indicated by the identification result, it is determined that the identification is correct this time. If any of the sample features indicated by the identification result does not match the target sample features, it is determined that the identification is incorrect this time. The feedback host 130 can feedback whether the identification result this time is in the feedback result. In this way, the running of the identification program and the control program can be verified based on the collected data and the sensing data.

[0077] It can be understood that, since the identification program is implemented based on an artificial intelligence algorithm, the identification accuracy of the program is a key performance indicator, and the feedback host 130 can count the running results of the identification program in the actual measurement process to calculate the accuracy of the identification program. In order to guide the attention of the trainee to the program running result, the feedback host 130 can display the running result of this time and the accuracy statistics of the identification program in each output feedback result.

[0078] Based on the above method, the feedback host 130 analyzes the theoretical running results of the identification program and the control program based on the real data obtained by the data collector 122 and the sensor 112d, and generates a feedback result to show the trainee, so as to provide analysis and statistics of the running situation of the identification program and the control program. The trainee can also intuitively and real-timely obtain the corresponding feedback information during the interesting measurement experiment, so that the trainee can pay attention to the program running situation while retaining the experiment interest, and further stimulate the interest of the trainee in program writing.

[0079] Those skilled in the art can understand that all or part of the steps in the above-mentioned implementation methods can be completed by instructing the relevant hardware by a program stored in a storage medium, including a plurality of instructions for causing a device (which can be a single-chip microcomputer, a chip, etc.) or a processor to execute all or part of the steps of the various embodiments of the present application. The aforementioned storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), a magnetic disk or an optical disk, and various media that can store program codes.

[0080] The above are preferred embodiments of the present application, and do not limit the protection scope of the present application, so that: all equivalent changes made according to the structure, shape, principle of the present application should be covered within the protection scope of the present application.

Claims

1. An intelligent sorting simulation teaching system, characterized in that, The system comprises a sorting pad, a sorting table and a feedback host; The sorting pad is divided into a sorting area and a plurality of storage areas, the sorting area is used for placing the sorting table, the sorting table is used for placing a target sample to be sorted, and the storage area is used for placing the target sample after sorting; the sorting table is provided with a data collector, each storage area comprises a hard layer and a flexible layer, the hard layer and the flexible layer are connected at the four peripheral edges to form a closed hollow structure, the hollow structure is filled with inert gas, and a stress sensor is arranged close to the flexible layer to collect the stress change of the flexible layer; when a sample is placed on the storage area, the gravity of the sample acts on the upper hard layer, and the deformation amount of the hard layer is very small; under the action of the gravity of the sample, the inert gas is subjected to extrusion, causing the deformation of the flexible layer, and the stress sensor can collect the stress change of the flexible layer; compared with being arranged close to the hard layer, the stress sensor arranged close to the flexible layer is more stable, and the influence of the stress sensor itself on the collection result can be reduced; the data collector and each sensor are signal-connected with the feedback host, and send collection data and sensing data to the feedback host based on the signal connection; The feedback host confirms the target sample characteristics of the target sample based on the collection data, determines the target storage area where the target sample is located according to the sensing data, and compares the target sample characteristics and the target storage area with a pre-established storage relationship to determine whether the sorting result of the target sample meets the expectation, and outputs a feedback result; After receiving the collection data, the feedback host saves it in a first queue and records the receiving time; after receiving the sensing data, the feedback host saves it in a second queue and records the receiving time; at the same time, the feedback host calculates the difference between the receiving time of the sensing data and the latest recorded receiving time of the collection data in the first queue; if the time difference meets a preset condition, the sensing data is determined as valid data, i.e. data generated by the system, wherein the preset condition includes that the time difference is consistent with the operating time length of the mechanical arm, and the operating time length of the mechanical arm is pre-collected and is the time consumed by the system when the sample is placed on the sorting table to the time when the sample is transferred to the storage area during normal operation; since the distances between the storage areas and the mechanical arm are different, the operating time length of the mechanical arm can take an interval value, and the two endpoints are the time consumed to the nearest storage area and the time consumed to the farthest storage area; if the time difference does not meet the preset condition, the sensing data is directly discarded, i.e. it is deleted from the second queue; at the same time, if no valid sensing data is received within a preset time length, the collection data is marked as invalid, and the feedback host displays the invalid data in the feedback result.

2. The system of claim 1, wherein, The system further comprises a mechanical arm and a mechanical arm controller, the mechanical arm is connected with the mechanical arm controller and fixed in an operation area of the sorting pad, the operation area is adjacent to the sorting area; the mechanical arm controller is programmed with a mechanical arm control program for controlling the mechanical arm to transfer the target sample from the sorting table to the corresponding storage area, wherein the control program determines the corresponding storage area of the target sample based on the storage relationship.

3. The system of claim 2, wherein, The system further comprises a sample identifier, the sample identifier runs an identification program based on an artificial intelligence algorithm, the sample identifier identifies the characteristics of the target sample based on the identification program and sends the identification result to the mechanical arm controller, and the mechanical arm controller determines the corresponding storage area of the target sample based on the identification result and the storage relationship.

4. The system of claim 3, wherein, The sample identifier is connected with the feedback host and sends the identification result to the feedback host; the feedback host compares the identification result with the target sample characteristics and generates and feeds back the accuracy rate of the identification program based on the comparison result.

5. The system of claim 1, wherein, The sorting pad further has an induction controller built-in and connected with each sensor through a signal line to receive the sensing data sent by each sensor, and after filtering each sensing data, the sensing data is sent to the feedback host.

6. The system of claim 1, wherein, The sorting table further comprises a base, the base comprises a first layer plate and a second layer plate arranged oppositely, the first layer plate is coincident with the center point of the second layer plate and has an area smaller than the second layer plate, the data collector is fixed to the first layer plate through a connecting structure, and the surface of the second layer plate away from the first layer plate is provided with an anti-skid structure.

7. The intelligent sorting simulation teaching method is characterized in that, The method is applied to the feedback host of claim 1, and the method comprises: Real-time receiving of sensing data sent by a sorting pad built-in sensor and collection data sent by a data collector; Confirmation of target sample characteristics of a target sample based on the collection data; Determination of a target storage area where the target sample is located according to the sensing data; Comparison of the target sample characteristics and the target storage area with a pre-established storage relationship to determine whether the sorting result of the target sample meets the expectation and output a feedback result.

8. The method of claim 7, wherein, The real-time receiving of sensing data sent by a sorting pad built-in sensor comprises: Real-time receiving of sensing data sent by an induction controller built-in the sorting pad, wherein the sensing data is collected by each sensor and obtained through filtering of the induction controller.

9. The method of claim 8, wherein, The filtering method comprises: Each sensor sends the collected initial sensing data to the induction controller in real time; The induction controller respectively compares each initial sensing data sent by each sensor with the last data; If the initial sensing data changes positively compared with the last data, the initial sensing data is taken as the filtered sensing data.

10. The method of claim 7, wherein, The determination of a target storage area where the target sample is located according to the sensing data comprises: Judgment of whether the sensing data is valid based on the difference between the receiving time of the collection data and the sensing data; If valid, determine the storage area where the sensor sending the sensor data is located as the target storage area.

Images