1. The utility model provides a waste classification device based on unmanned vehicle which characterized in that includes:

an unmanned trolley;

sorter, set up in on the unmanned vehicle, sorter includes:

the cover body comprises a first through hole, and the first through hole is used for placing garbage;

the control module is accommodated in the cover body;

the detection bin is accommodated in the cover body and comprises an installation plate, an accommodating body, a sensor module, a motor module and a tray, wherein a second through hole is formed in the installation plate, the accommodating body comprises an accommodating space which is communicated up and down, the first through hole is communicated with the accommodating space, the second through hole is positioned below the accommodating space, the sensor module is used for detecting garbage in the accommodating space, the sensor module and the motor module are both connected with the control module, the motor module comprises a first motor, the first motor is connected with the tray, and the first motor is used for controlling the movement of the tray so as to control the communication state between the accommodating space and the second through hole;

the storage cabinet comprises a guide structure, a second motor and a plurality of storage spaces, wherein the storage spaces are used for containing different types of garbage respectively, the guide structure is located below the second through hole and connected with the second motor, the second motor is connected with the control module, and the second motor is used for controlling the direction of the guide structure so as to guide the garbage to enter the type of the garbage corresponding to the storage spaces.

2. The waste sorting device of claim 1, further comprising: the sensor module comprises an inductive proximity switch sensor, a capacitive proximity switch sensor, a diffuse reflection photoelectric switch sensor, an ultrasonic sensor and a pressure sensor, wherein the pressure sensor is positioned on the tray.

3. The waste sorting device of claim 1, further comprising: the motor module further comprises a third motor, wherein the third motor is connected with the sensor module and used for controlling the movement of the sensor module so as to adjust the distance between the sensor module and the accommodating space.

4. The waste sorting device of claim 1, further comprising: the unmanned trolley comprises a sensing module and a core processing module, wherein the sensing module is connected with the core processing module, the sensing module is used for acquiring environmental information and positioning information, and the core processing module is used for controlling the unmanned trolley according to the environmental information and the positioning information.

5. A sorting method applied to the garbage sorting device according to any one of claims 1 to 4, comprising:

the unmanned trolley moves to a target position;

acquiring sensor data of current garbage through a sensor module; the sensor data includes detection data of the ultrasonic sensor and a set of distances including a first trigger distance of the inductive proximity switch sensor, a second trigger distance of the capacitive proximity switch sensor, and a third trigger distance of the diffuse reflectance optoelectronic switch sensor;

determining weights corresponding to the detection data, the first trigger distance, the second trigger distance and the third trigger distance respectively according to a preset database; the preset database comprises sensor sample data corresponding to a plurality of types of garbage;

according to the weight and the sensor sample data, determining the mean square error of the current garbage and each sensor sample data;

determining a classification result of the current garbage according to a preset threshold and the mean square error;

and according to the classification result, guiding the current garbage to enter the storage space of the corresponding type through a guide structure.

6. The classification method according to claim 5, characterized in that: the sensor module includes a pressure sensor and a position sensor, the method further comprising:

determining first time when the pressure sensor acquires a stable pressure value of the current garbage and second time when the current garbage continuously triggers the in-place sensor;

and when the first time is greater than or equal to a first threshold value and the second time is greater than or equal to a second threshold value, the step of acquiring the sensor data of the current garbage through the sensor module is carried out.

7. The classification method according to claim 5, characterized in that: the mounting panel has a sensor track, the sensor module movably set up in the sensor track, acquire the sensor data of current rubbish through the sensor module, include:

controlling the sensor module to move on the sensor track at a preset speed, and recording the movement starting time and the trigger time of each sensor;

and determining the distance set according to the preset speed, the movement starting time and the trigger time.

8. The classification method according to claim 5, characterized in that: the determining the classification result of the current garbage according to the preset threshold and the mean square error comprises:

when the minimum mean square error is smaller than or equal to the preset threshold, taking the garbage type corresponding to the minimum mean square error as the classification result of the current garbage;

alternatively, the first and second electrodes may be,

when the minimum mean square error is larger than the preset threshold, executing one of the following steps:

prompting the failure of identification;

or, returning to the step of obtaining the sensor data of the current garbage through the sensor module until the minimum mean square error is less than or equal to the preset threshold, and taking the garbage type corresponding to the minimum mean square error as the classification result of the current garbage;

alternatively, the current waste is discharged out of the waste sorting device.

9. The classification method according to claim 5, characterized in that: after the step of leading current rubbish to get into the storing space of corresponding kind through guide structure, still include:

entering a user money collection stage;

determining the payment state of a user in a payment collection stage of the user;

and when the payment state is successful, ending the user payment stage.

10. The classification method according to claim 5, characterized in that: the unmanned vehicle moving to the target position includes:

when the unmanned trolley is in a moving state and when the classification device receives garbage or an object approaches the unmanned trolley, determining the current position of the unmanned trolley as a target position;

alternatively, the first and second electrodes may be,

determining a target position by receiving a user instruction to move the unmanned vehicle to the target position.

Background

Along with the development of society, the degree of urbanization is deepened continuously, and the requirement of the society on environmental protection is higher and higher nowadays, for example, garbage classification can well improve the sanitary state of our living environment by classifying garbage, and is beneficial to improving the efficiency of subsequent targeted treatment on garbage. At present, the classification of garbage mainly depends on the cooperation of a recycling product provider, and the material of a recycling product is identified in an intelligent or semi-intelligent mode by equipment deployed in a recycling station, so that the garbage is classified. Various problems exist in the intellectualization of the classification and recovery of the garbage at the present stage, for example: 1. the consciousness and the ability of the classification of residents are deficient, and the resistance of popularizing the garbage classification is large; 2. a large amount of investment is needed for deploying a large amount of intelligent garbage classification and recovery equipment; 3. the single park needs to be covered completely, the equipment investment is very large, and the power consumption is large; 4. the classification and identification of the recycled products need to be matched with users, and the reliability of identification means is low.

Disclosure of Invention

In view of the above, in order to solve the above technical problems, the present invention provides a garbage sorting device and a sorting method based on an unmanned vehicle, which are used for improving the convenience and accuracy of garbage sorting and reducing the cost.

The technical scheme adopted by the invention is as follows:

a waste classification device based on unmanned trolley comprises:

an unmanned trolley;

sorter, set up in on the unmanned vehicle, sorter includes:

the cover body comprises a first through hole, and the first through hole is used for placing garbage;

the control module is accommodated in the cover body;

the detection bin is accommodated in the cover body and comprises an installation plate, an accommodating body, a sensor module, a motor module and a tray, wherein a second through hole is formed in the installation plate, the accommodating body comprises an accommodating space which is communicated up and down, the first through hole is communicated with the accommodating space, the second through hole is positioned below the accommodating space, the sensor module is used for detecting garbage in the accommodating space, the sensor module and the motor module are both connected with the control module, the motor module comprises a first motor, the first motor is connected with the tray, and the first motor is used for controlling the movement of the tray so as to control the communication state between the accommodating space and the second through hole;

the storage cabinet comprises a guide structure, a second motor and a plurality of storage spaces, wherein the storage spaces are used for containing different types of garbage respectively, the guide structure is located below the second through hole and connected with the second motor, the second motor is connected with the control module, and the second motor is used for controlling the direction of the guide structure so as to guide the garbage to enter the type of the garbage corresponding to the storage spaces.

Further, the sensor module includes an inductive proximity switch sensor, a capacitive proximity switch sensor, a diffuse reflectance electro-switch sensor, an ultrasonic sensor, and a pressure sensor, the pressure sensor being located on the tray.

Further, the motor module further comprises a third motor, and the third motor is connected with the sensor module and used for controlling the movement of the sensor module so as to adjust the distance between the sensor module and the accommodating space.

Further, the accommodating body is provided with a plurality of hollow structures, and the hollow structures are communicated with the accommodating space and are correspondingly arranged with the sensor module.

Furthermore, the unmanned trolley comprises a sensing module and a core processing module, wherein the sensing module is connected with the core processing module, the sensing module is used for acquiring environmental information and positioning information, and the core processing module is used for controlling the unmanned trolley according to the environmental information and the positioning information.

The invention also provides a classification method applied to the garbage classification device, which comprises the following steps:

the unmanned trolley moves to a target position;

acquiring sensor data of current garbage through a sensor module; the sensor data includes detection data of the ultrasonic sensor and a set of distances including a first trigger distance of the inductive proximity switch sensor, a second trigger distance of the capacitive proximity switch sensor, and a third trigger distance of the diffuse reflectance optoelectronic switch sensor;

determining weights corresponding to the detection data, the first trigger distance, the second trigger distance and the third trigger distance respectively according to a preset database; the preset database comprises sensor sample data corresponding to a plurality of types of garbage;

according to the weight and the sensor sample data, determining the mean square error of the current garbage and each sensor sample data;

determining a classification result of the current garbage according to a preset threshold and the mean square error;

and according to the classification result, guiding the current garbage to enter the storage space of the corresponding type through a guide structure.

Further, the sensor module includes a pressure sensor and a position sensor, and the method further includes:

determining first time when the pressure sensor acquires a stable pressure value of the current garbage and second time when the current garbage continuously triggers the in-place sensor;

and when the first time is greater than or equal to a first threshold value and the second time is greater than or equal to a second threshold value, the step of acquiring the sensor data of the current garbage through the sensor module is carried out.

Further, the mounting plate has a sensor track, the sensor module is movably disposed on the sensor track, the sensor data of the current garbage is obtained through the sensor module, including:

controlling the sensor module to move on the sensor track at a preset speed, and recording the movement starting time and the trigger time of each sensor;

and determining the distance set according to the preset speed, the movement starting time and the trigger time.

Further, the determining the classification result of the current garbage according to the preset threshold and the mean square error includes:

when the minimum mean square error is smaller than or equal to the preset threshold, taking the garbage type corresponding to the minimum mean square error as the classification result of the current garbage;

alternatively, the first and second electrodes may be,

when the minimum mean square error is larger than the preset threshold, executing one of the following steps:

prompting the failure of identification;

or, returning to the step of obtaining the sensor data of the current garbage through the sensor module until the minimum mean square error is less than or equal to the preset threshold, and taking the garbage type corresponding to the minimum mean square error as the classification result of the current garbage;

alternatively, the current waste is discharged out of the waste sorting device.

Further, after the step of guiding the current garbage to enter the storage space of the corresponding type through the guiding structure, the method further comprises the following steps:

entering a user money collection stage;

determining the payment state of a user in a payment collection stage of the user;

and when the payment state is successful, ending the user payment stage.

Further, the unmanned vehicle moving to a target position includes:

when the unmanned trolley is in a moving state and when the classification device receives garbage or an object approaches the unmanned trolley, determining the current position of the unmanned trolley as a target position;

alternatively, the first and second electrodes may be,

determining a target position by receiving a user instruction to move the unmanned vehicle to the target position.

The invention has the beneficial effects that: when garbage enters the accommodating space through the first through hole, the garbage in the accommodating space is detected through the sensor module and is transmitted to the control module for processing, so that the control module can judge the type of the garbage, then the tray moves by controlling the first motor, the garbage can enter the storage cabinet through the second through hole, and then the direction of the guide structure is controlled through the second motor to guide the garbage to enter the storage space of the type corresponding to the garbage, so that the garbage classification is well realized, and the accuracy of the garbage classification is improved; and set up in unmanned dolly at sorter, for place the garbage bin in fixed position, can improve the convenience of waste classification, it is more intelligent, simultaneously for the scheme that has the total coverage big input, can reduce the consumption and save the cost.

Drawings

FIG. 1 is a schematic view of a waste sorting device according to the present invention;

FIG. 2 is a schematic view of the sorting apparatus of the present invention;

FIG. 3 is a schematic diagram of a detection chamber according to an embodiment of the present invention;

FIG. 4 is a schematic view of a storage cabinet according to an embodiment of the present invention;

FIG. 5 is a schematic diagram of the classification method according to the present invention;

FIG. 6 is a flowchart of a classification method according to an embodiment of the present invention.

Detailed Description

In order to make the technical solutions better understood by those skilled in the art, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are only partial embodiments of the present application, but not all embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.

The terms "first," "second," "third," and "fourth," etc. in the description and claims of this application and the accompanying drawings are used for distinguishing between different objects and not for describing a particular order. Furthermore, the terms "include" and "have," as well as any variations thereof, are intended to cover non-exclusive inclusions. For example, a process, method, system, article, or apparatus that comprises a list of steps or elements is not limited to only those steps or elements listed, but may alternatively include other steps or elements not listed, or inherent to such process, method, article, or apparatus.

Reference herein to "an embodiment" means that a particular feature, structure, or characteristic described in connection with the embodiment can be included in at least one embodiment of the application. The appearances of the phrase in various places in the specification are not necessarily all referring to the same embodiment, nor are separate or alternative embodiments mutually exclusive of other embodiments. It is explicitly and implicitly understood by one skilled in the art that the embodiments described herein can be combined with other embodiments.

As shown in fig. 1, an embodiment of the present invention provides a garbage classification apparatus based on a drone vehicle, which includes a drone vehicle 100 and a classification apparatus 200. Specifically, the sorting device 200 is installed on the unmanned vehicle 100 and fixed by a fixing groove (not shown).

In the embodiment of the present invention, the classification apparatus 200 includes: lid 1, detection storehouse 2, locker 3 and control module A.

As shown in fig. 2, in the embodiment of the present invention, the cover 1 is provided with a first through hole 11 and an installation area 12, the first through hole 11 is used for placing garbage, and the installation area 12 has an interactive module (not shown). Optionally, the interaction module is connected with the control module and the background server, and comprises a plurality of indicator lights, an operable screen and a voice module, wherein the interaction module comprises three indicator lights as an example, and respectively represents three states of ready, running and abnormal; the operable screen is used for information prompt, two-dimensional code display and simple operation; the voice module is used for playing voice and prompting a user to know operation steps and garbage classification results conveniently.

In the embodiment of the present invention, the control module is housed in the cover body 1, and it needs to perform tasks of three aspects of control, communication and operation, and has a power supply, a processing chip, an ADC analog-to-digital conversion channel, a timer, a basic level input/output, a serial communication, a PWM, an auxiliary module, and the like. The auxiliary module comprises a mobile network module, SDRAM data records, a TFTLCD display screen, an electric frequency conversion circuit, an isolation and protection circuit and the like. It should be noted that the control module receives signals of various sensors, processes data received by the sensors through a preset algorithm, and controls the motors to complete the garbage classification process.

Optionally, the processing chip must have functions of a multi-channel ADC analog-to-digital conversion module, GPIO basic input and output, UART serial interface communication, second and millisecond timers, external interruption, PWM pulse width modulation, and the like, and can adapt to outdoor temperature and humidity conditions, operate stably for a long time, and have low power consumption, so that the model of STM32F407 is selected. In the embodiment of the invention, the GPIO is mainly used for outputting and controlling various switches in equipment, such as a relay, an indicator light, a buzzer and the like, and the display content of the LCD screen can be controlled by matching a plurality of IO outputs. In order to improve the sensitivity and the real-time performance, the input switching signal is mainly interrupted by external processing, for example, when the time of the triggering signal of the switching sensor is recorded, the recording is required to be more accurate (the identification accuracy is influenced).

The core processor is driven to run by mainly generating pulse frequency signals which can be accepted by the stepping motor through PWM. The higher the frequency of the input, the faster the motor moves. Currently the speed of each motor is constant, i.e. the pulse frequency output by the core processor is constant. The UART serial port communication is used for carrying out control operation with other expansion modules, such as wifi and voice modules. In addition, the TF card is read and written by using the SPI interface, and a simple FAT32 file system needs to be established at the same time, so that data needing to be stored in a power-off mode is backed up in a file form. Except for the switch type sensor, the pressure sensor and the ultrasonic wave are all output as analog signals. STM32 multichannel ADC analog-to-digital conversion can monitor the analog signals of a plurality of sensors simultaneously, and combines ADC interruption and DMA, can more accurately save denser discrete data record. And finally, taking out the cache data in the calculation stage for operation. Timers, including a millisecond timer and a second timer, each start interrupt accumulation at system start-up, and other programs may obtain count values as needed.

As shown in fig. 2 and 3, in the embodiment of the present invention, the detecting chamber 2 is accommodated in the cover body 1, and includes an installation plate 21, an accommodating body 22, a tray 23, a sensor module, and a motor module.

As shown in fig. 2 and 3, optionally, both the sensor module and the motor module are connected with the control module, the sensor module including an inductive proximity switch sensor 41, a capacitive proximity switch sensor 42, a diffuse reflective photoelectric switch sensor 43, an ultrasonic sensor 44, a pressure sensor (not shown) and a position sensor (not shown). Wherein the pressure sensor is located in the tray 23.

As shown in fig. 2 and 3, the mounting plate 21 may be provided with a second through hole (not shown), a moving rail 211, and a sensor rail 212. Wherein, the tray 23 is movably disposed on the moving track 211, and the sensor module is movably disposed on the sensor track 212. It should be noted that the sensor module is movably disposed on the sensor track 212, which means that at least one sensor in the sensor module is movably disposed on the sensor track 212, in the embodiment of the present invention, the inductive proximity switch sensor 41, the capacitive proximity switch sensor 42, and the diffuse reflection photoelectric switch sensor 43 are movably disposed on the sensor track 212, the ultrasonic sensor 44 is disposed on the mounting plate 21 on one side of the accommodating body 22, and the position sensor is disposed on one side of the accommodating body 22. Optionally, the longest distance of the sensor track 212 is 25 cm.

As shown in fig. 2 and fig. 3, optionally, the accommodating body 22 includes an accommodating space 221 penetrating up and down and a plurality of hollow structures 222. The first through hole 11 is communicated with the accommodating space 221, the second through hole is located below the accommodating space 221, and the sensor module is used for detecting garbage in the accommodating space 221. The hollow structure 222 is connected to the accommodating space 221 and is disposed corresponding to the sensor module. It should be noted that, the hollow-out structures 222 and the sensor modules are arranged correspondingly, that is, at least one sensor in the sensor modules faces the hollow-out structures 222, so that the sensor can acquire data of the garbage in the accommodating space 221 through the hollow-out structures 222, in the embodiment of the present invention, the inductive proximity switch sensor 41, the capacitive proximity switch sensor 42, the diffuse reflection photoelectric switch sensor 43, and the ultrasonic sensor 44 correspond to one hollow-out structure 222, respectively, so that each sensor acts on the garbage at a proper angle. In the embodiment of the present invention, by providing multiple types of sensors, information such as a position, a weight, and a material property of the garbage can be detected, so that the control module can perform recognition processing to classify the garbage.

It should be noted that the inductive proximity switch sensor 41 is very sensitive to metal (including aluminum), and when the recycled product is a pop can or other metal material, the sensor is activated very quickly, i.e., the activation distance is large. The capacitive proximity switch sensor 42 is effective for all materials including metal, glass, plastic, wood, and the sensor excitation distance may vary depending on the material. The diffuse reflection photoelectric switch sensor 43 is inferior to metal, mirror surface, and paper in effect to transparent articles such as white plastic and transparent glass, and therefore can be one of the main bases for identifying the material of garbage. Any material of the ultrasonic sensor 44 can absorb sound, but the degree of absorption is very different, so according to the characteristic, the sound absorption performance of the recycled product is monitored by matching the ultrasonic transmitter and the receiver. The pressure sensor can acquire the weight of the garbage, and the material range of the recycled product or whether the recycled product is filled with sundries or not can be roughly known according to the weight by matching with data parameters of other sensors.

As shown in fig. 2 and 3, the motor module optionally includes a first motor 51 and a third motor 52. The first motor 51 is connected with the tray 23, and the first motor 51 is used for controlling the movement of the tray 23 to control the communication state between the accommodating space 221 and the second through hole; the third motor 52 is connected to the sensor module for controlling the movement of the sensor module on the sensor rail 212 to adjust the distance between the sensor module and the receiving space 221. It should be noted that, the initial position of the tray 23 is located on the moving track 211 and below the accommodating body 22, i.e. between the accommodating space 221 and the second through hole, the communication state is "closed", and after the tray 23 moves on the moving track 211 from the initial position, the communication state between the accommodating space 221 and the second through hole is "communicated", and at this time, the garbage in the accommodating space 221 can enter the storage cabinet 3 through the second through hole.

As shown in fig. 2 and 4, the storage cabinet 3 may alternatively include a guiding structure 31, a second motor 32 and a plurality of storage spaces 33, wherein the plurality of storage spaces 33 are respectively used for placing different types of garbage, and each storage space 33 may store one type of garbage, or each two storage spaces 33 may store one type of garbage, without limitation. For example, the types of the garbage respectively corresponding to the four storage spaces 33 are glass, plastic, metal, and others. The guiding structure 31 is located below the second through hole and connected with the second motor 32, the second motor 32 is connected with the control module, and the second motor 32 is used for controlling the rotating direction of the guiding structure 31 to guide the garbage to enter the storage space 33 of the type corresponding to the garbage. Specifically, after the control module analyzes the sensor data acquired by the sensor module to determine the type of the garbage, the control module controls the guiding structure 31 to guide the garbage to enter the storage space 33 of the type corresponding to the garbage. It will be appreciated that the guide structure 31 has a cushioning effect on the waste in addition to a guiding effect.

Optionally, a compressor, a shock absorber or a packaging device can be arranged below or at the side of the storage cabinet 3, for example, the pop can needs to be flattened during storage and is not too space-consuming; the shock absorber is used for preventing the glass bottle from being broken when falling; the packaging equipment is convenient for recovery personnel to collect in order to arrange the recovered materials.

Optionally, the unmanned vehicle comprises a sensing module, a core processing module and a communication module, wherein the sensing module is connected with the core processing module, and the communication module is connected with the core processing module, the server and the control module and used for receiving instructions and feedback information. Optionally, the communication module communicates with the classification device via a CAN bus.

Specifically, the sensing module comprises a front binocular camera and a rear binocular camera, a front laser radar, a rear laser radar, eight millimeter wave radars and a differential positioning module, wherein the front binocular camera and the rear binocular camera are used for imaging an obstacle in front of the unmanned trolley and identifying a marked scenery; the front and the back laser radars respectively obtain the far and near depths of the barrier, namely ranging; the periphery of the trolley is provided with eight millimeter wave radars which are used for detecting objects close to the periphery of the trolley body; and the differential GPS positioning module is used for acquiring the differential positioning module, so that the walking route can be conveniently planned. It should be noted that the relevant information, namely the environmental information, the environmental information and the positioning information obtained by the front and rear two binocular cameras, the front and rear two laser radars and the eight millimeter wave radars are transmitted to the core processing module for processing,

the system comprises a core processing module, a vehicle-mounted control module and a vehicle-mounted control module, wherein the vehicle-mounted control module is used for controlling the electronic driving direction, braking and an accelerator; and the trolley control module is used for calculating the current running environment according to the instruction sent by the classification device and the environment information and the positioning information, planning a route and coordinating the trolley control module to control the unmanned trolley. Alternatively, the instruction issued by the recycling device may be an instruction input through the interaction module or an instruction issued when the sorting device receives garbage, or an instruction received by the sorting device from a server, or the like.

It should be noted that the unmanned vehicle provided by the embodiment of the invention has an L4-level automatic driving capability, and a map is built in a core processing module, so that functions of low-speed cruising, obstacle avoidance, situation perception and the like can be realized. The user can obtain the position state information of all the garbage classification devices through a mobile phone application program, a WeChat applet or an APP, and can also reserve a certain classification device loaded on an unmanned trolley to go to a certain place (target position) on a specified line for garbage recycling service.

As shown in fig. 5, an embodiment of the present invention further provides a classification method applied to the above garbage classification apparatus, including steps S100 to S600:

s100, the unmanned trolley moves to a target position.

Optionally step 2100 comprises step S101 or step S102:

s101, when the unmanned trolley is in a moving state and when the classification device receives garbage or an object approaches the unmanned trolley, determining the position of the current unmanned trolley as a target position.

Specifically, the unmanned vehicle has two service modes, namely a fixed service mode and a mobile service mode, and the service modes can be controlled by a server or a classification device. When the unmanned trolley is in a mobile service mode, the server sends a 'tracing instruction' through a preset route to control the unmanned trolley to move according to the preset route to execute a tracing task, and when the unmanned trolley is in the moving state, the classifying device receives garbage input from a user, the current position of the unmanned trolley is determined as a target position, the unmanned trolley moves to the target position and stays in the 'service state', and the unmanned trolley exits the 'service state' until garbage classification is completed. And when the unmanned trolley is in a fixed service mode, the unmanned trolley stops at a fixed position to wait for a user to throw garbage. Alternatively, when the unmanned vehicle is in a moving state and an object (e.g., a person) approaches the unmanned vehicle (e.g., the distance from the unmanned vehicle is less than an approach threshold), the unmanned vehicle may take the current position as the target position and then enter a "pause state" to notify the classification device of the information, and the classification device starts timing. If the sorting device does not enter the waste recycling workflow (namely, the garbage is not received) within the timing threshold (for example, 30 seconds), a 'tracing instruction' or a 'position instruction' is sent to the unmanned trolley, namely, the unmanned trolley enters the flow service mode or the target task mode, the unmanned trolley is enabled to avoid obstacles and leave, and if the waste recycling workflow (namely, the garbage is received) is triggered within the timing threshold (for example, 30 seconds), the sorting device sends a 'service state' instruction, and the trolley enters the 'service state'.

And S102, determining a target position by receiving a user instruction so that the unmanned trolley moves to the target position.

Specifically, the user instruction can be applied to the server by the user through modes such as APP, public numbers and the like, the server generates the user instruction and sends the user instruction to the classification device and the unmanned trolley, and therefore the target task mode is entered. For example, when a user applies through an APP, a server sends a user instruction including a user position (i.e., a target position) to a classification device and an unmanned vehicle according to the application and the position (i.e., the target position) of the user, a core processing module of the unmanned vehicle generates a planned route and controls the unmanned vehicle to move to the target position according to the planned route, the unmanned vehicle stays at the target position for a preset stay time (e.g., 30 seconds), the user is waited to throw garbage into the unmanned vehicle, if the unmanned vehicle operates within the preset stay time, the unmanned vehicle automatically enters a fixed service mode or a mobile service mode, and if the user instruction includes a plurality of target positions, the core processing module is responsible for completing optimal path selection and preferentially reaches one of the target positions.

Optionally, the server or the classification device queries the state of the unmanned vehicle at regular time, and the control module of the classification device sends query information to the classification device through the CAN bus every 2 seconds to know various information on the classification device, including: an operation state, a position (corrected GPS information), a power supply state, a vehicle body abnormality, and the like, and the operation is continued without being interfered by other operations. Optionally, when the garbage classification device is in an abnormal maintenance state, for example, when the classification device or the unmanned vehicle has a fault, including but not limited to abnormal information such as too low battery voltage, the server may stop the service itself, the server may send a "return command" to the unmanned vehicle, and the unmanned vehicle may return to the maintenance point according to a preset fixed path.

S200, acquiring sensor data of the current garbage through a sensor module.

In an embodiment of the invention, the sensor data comprises detection data of the ultrasonic sensor and a distance set, wherein the distance set comprises a first trigger distance of the inductive proximity switch sensor, a second trigger distance of the capacitive proximity switch sensor and a third trigger distance of the diffuse reflectance optoelectronic switch sensor.

In order to identify the recycled product in a non-visual manner, it is necessary to consider the characteristics of the material, and the material may have different properties, such as sound wave absorption, light reflection, dielectric constant, conductivity, and weight. The single attribute information of the detection objects with different materials may be similar, but the difference can be highlighted by fusing a plurality of information. Therefore, based on the multi-information difference, different types of sensors are combined, and the garbage classification result can be obtained through the analysis of the difference of the trigger distances of the different types of sensors.

Optionally, step S200 includes the steps of: S110-S120:

and S110, controlling the sensor module to move on the sensor track at a preset speed, and recording the movement starting time and the trigger time of each sensor.

Optionally, the sensor module is arranged at an initial position, so that the inductive proximity switch sensor, the capacitive proximity switch sensor and the diffuse reflection photoelectric switch sensor of the sensor module cannot detect the garbage in the accommodating space at the initial position, and the sensor module is controlled to move, specifically, the inductive proximity switch sensor, the capacitive proximity switch sensor and the diffuse reflection photoelectric switch sensor are controlled to move at a preset speed on the sensor track, the triggering time of different sensors is different due to the difference of garbage types in the moving process, and the triggering time of each sensor is recorded at the moment. It should be noted that, when the sensor module starts to move, the time at this time, i.e., the movement start time, is recorded. In addition, the preset speed can be set according to actual needs and is not limited.

And S120, determining a distance set according to the preset speed, the movement starting time and the trigger time.

Specifically, the formula for calculating the trigger distance is as follows: the triggering distance is [ total length of the sensor track- (triggering time-movement starting time) × preset speed ], and the first triggering distance of the inductive proximity switch sensor, the second triggering distance of the capacitive proximity switch sensor and the third triggering distance of the diffuse reflection photoelectric switch sensor can be obtained through calculation according to the triggering time of the inductive proximity switch sensor, the capacitive proximity switch sensor and the diffuse reflection photoelectric switch sensor. It should be noted that the detection data of the ultrasonic sensor can be obtained by detecting the garbage with the ultrasonic sensor. In some embodiments, the trigger distance may also be (trigger time-movement start time) a preset speed, and is not particularly limited.

S300, determining weights corresponding to the detection data, the first trigger distance, the second trigger distance and the third trigger distance respectively according to a preset database.

Specifically, the preset database includes sensor sample data corresponding to a plurality of types of garbage, and each type of sensor sample data also includes sample detection data of the ultrasonic sensor and sample trigger distances of the inductive proximity switch sensor, the capacitive proximity switch sensor and the diffuse reflection photoelectric switch sensor. The sensor sample data may be learned by a machine learning method, so as to determine weights corresponding to the sample trigger distances of the sensor sample data and the inductive proximity switch sensor, the capacitive proximity switch sensor, and the diffuse reflection photoelectric switch sensor in the sensor sample data, then the weight of the sample detection data is used as the weight of the detection data, specifically a first weight, the weight of the sample trigger distance of the inductive proximity switch sensor is used as the weight of the first trigger distance, specifically a second weight, the weight of the sample trigger distance of the capacitive proximity switch sensor is used as the weight of the second trigger distance, specifically a third weight, and the weight of the sample trigger distance of the diffuse reflection photoelectric switch sensor is used as the weight of the third trigger distance, specifically a fourth weight.

For example, set of sensor sample data S₁，S₂，...，S_n，S_nThe nth sensor sample data is represented, and the sensor sample data set contains sensor sample data of different types of garbage, for example, the garbage sample data in the embodiment of the inventionThe category of (b) includes glass, plastic, metal, and others, i.e., n is 4 or greater. In addition, each S_i(i.e. ith sensor sample data) has a sensor data set comprising p measured data σ collected by different types of sensors_i1，σ_i2，...，σ_ip(i.e., sample detection data and individual sample detection data), σ_ipFor the p-th measured data in the ith sensor sample data, for example, the invention includes an inductive proximity switch sensor, a capacitive proximity switch sensor, a diffuse reflectance photoelectric switch sensor and an ultrasonic sensor, i.e., p is 4, it is understood that when the variety of sensors utilized increases, the value of p increases accordingly. Then, determining a weight set w corresponding to the measured data by using a sensor sample data set₁，w₂，...，w_p，w_pRepresenting the weight corresponding to the p measured data. In the embodiment of the present invention, the example of the embodiment of the present invention is described as including four measured data, that is, the first weight is w₁The second weight is w₂The third weight is w₃The fourth weight is w₄。

S400, determining the mean square error of the current garbage and each sensor sample data according to the weight and the sensor sample data.

Calculating the mean square error delta of the current garbage and the sample data of each sensor_i：

Wherein, X_jIs the jth, X_jData acquired for the jth sensor,. sigma_ijIs the jth measured data, w, in the ith sensor sample data_jThe weight corresponding to the jth measured data (i.e. the weight corresponding to the data acquired by the jth sensor). For example, the sensor data acquired by the sensor module is X₁，X₂，...，X_p，X_pFor the data acquired by the p-th sensor, in the embodiment of the invention, X₁To detect data, X₂Is the first touchDistance of hair, X₃Is two trigger distances, X₄Is the third trigger distance. Determining a weighted square error, X, by a formula₁，X₂，X₃，X₄Respectively corresponding weights are w₁，w₂，w₃，w₄. It should be noted that the mean square error δ between the current garbage and the sample data of each sensor can be calculated by the above formula_iObtaining a mean square error set delta₁，δ₂，...，δ_i。

S500, determining the classification result of the current garbage according to a preset threshold and the mean square error.

Optionally, step S500 includes the steps of: s410 or S420:

and S410, when the minimum mean square error is smaller than or equal to a preset threshold value, taking the garbage type corresponding to the minimum mean square error as a current garbage classification result.

Specifically, the minimum mean square error in the mean square error set is selected as a target mean square error epsilon, when epsilon is less than or equal to a preset threshold value K, sensor sample data corresponding to the minimum mean square error is determined, the type of garbage corresponding to the sensor sample data is used as a current garbage classification result, for example, when the type of garbage corresponding to the sensor sample data corresponding to the minimum mean square error is glass, the glass is used as the current garbage classification result.

S420, when the minimum mean square error is larger than a preset threshold, executing one of the steps S421, S422 or S423:

and S421, prompting that the identification fails.

Specifically, when epsilon is larger than a preset threshold value K, namely the current garbage classification identification fails, the user identification failure is prompted on the operable screen.

S422, returning to the step of acquiring the sensor data of the current garbage through the sensor module until the minimum mean square error is less than or equal to a preset threshold value, and taking the garbage type corresponding to the minimum mean square error as the classification result of the current garbage;

specifically, when epsilon is larger than the preset threshold K, that is, the current garbage classification identification fails, because there is a possibility that accidental factors may affect the classification result, at this time, re-detection may be performed, and the step S100 is returned to the step S410.

And S423, discharging the current garbage out of the garbage classification device.

Specifically, when epsilon is larger than a preset threshold value K, that is, the current garbage classification identification fails, no sample of the current garbage category exists in the preset database, and the sample exceeds the classification range of the current device, and at this time, the current garbage is discharged out of the garbage classification device.

S600, according to the classification result, guiding the current garbage to enter the storage space of the corresponding type through a guide structure.

Specifically, after the classification result is obtained, the rotation direction of the guide structure is controlled to guide the current garbage to enter the storage space of the corresponding type. For example, after the classification result of the current garbage is obtained as glass, the rotation direction of the guide structure is controlled to guide the current garbage to enter the storage space correspondingly accommodating the glass.

Optionally, the classification method according to the embodiment of the present invention further includes: S601-S602:

s601, determining first time when the pressure sensor acquires a stable pressure value of the current garbage and second time when the current garbage continuously triggers the in-place sensor.

In the embodiment of the invention, a first time when the pressure sensor acquires a stable pressure value of the current garbage and a second time when the current garbage continuously triggers the in-place sensor are determined, that is, when the pressure sensor acquires the stable pressure value of the current garbage (for example, within a first preset fluctuation range), timing is started to obtain the first time, and when the in-place sensor triggers, timing is started to obtain the second time.

S602, when the first time is greater than or equal to a first threshold value and the second time is greater than or equal to a second threshold value, acquiring the sensor data of the current garbage through the sensor module.

In the embodiment of the present invention, the first threshold and the second threshold are both 2 seconds, and it can be understood that the first threshold and the second threshold may be set according to actual needs. Specifically, when the first time is 2 seconds or more and the second time is 2 seconds or more, step S100 is performed. It should be noted that the sensor module according to the embodiment of the present invention is provided with sensor data and an in-place sensor, so that validity and authenticity of the sensor data acquired by the sensor module can be ensured. Optionally, when the first time is greater than or equal to the first threshold and the second time is greater than or equal to the second threshold, the garbage classification device enters an operation stage, and the "operation" indicator light is turned on, otherwise, an abnormal alarm is given.

Optionally, step S600 is followed by the step of: S701-S703:

and S701, specifically, if the classification result of the current garbage is successfully identified, the classification result is uploaded to a server at the entering stage so that the server can calculate the cost according to the classification result.

S702, determining the payment state of the user in the user payment stage.

Specifically, after the cost is determined, a merchant account transfer related flow record and a merchant account transfer related flow record are generated, the two-dimensional codes corresponding to the related links are displayed on a controllable screen, a user can obtain a 'gold coupon' by scanning the two-dimensional codes, the user can exchange corresponding real or virtual currency through the 'gold coupon', when the user successfully obtains the 'gold coupon' by scanning the two-dimensional codes, the payment state is successful, and otherwise, the payment state is abnormal.

And S703, when the payment state is successful, ending the user payment stage.

Specifically, when the payment state is successful, the user collection stage is ended, and the garbage classification device returns to the standby state to wait for the input of garbage. Optionally, when the payment state is an abnormal state, recording abnormal data, and uploading the abnormal data to the server.

As shown in fig. 6, the method of the embodiment of the present invention is described in detail, and specifically includes the following steps (the execution sequence is not unique):

s901, assuming that when the unmanned vehicle has moved to the target position, when the garbage sorting device is in a standby state, it is determined whether the articles (garbage) enter the testing bin (detection bin), if so, it is waited for the articles to be stable, specifically, it may be determined whether the articles (garbage) are stable through steps S601-S602, and if not, it is continuously detected whether the articles (garbage) enter the testing bin (detection bin).

S902, after the articles are stabilized, pressure data a are recorded through a pressure sensor on a tray (carrying tray), then the garbage classification device enters an operation state, and ultrasonic scanning is carried out through an ultrasonic sensor to obtain detection data b which are recorded.

And S903, driving a third motor (a sensor group tray motor) through a control module to enable the sensor module to move, triggering the inductive proximity switch sensor, the capacitive proximity switch sensor and the diffuse reflection photoelectric switch sensor in the moving process of the sensor module, calculating and recording corresponding excitation distance data c, d and e (namely each triggering distance) through the control module, and executing the next step after detection is finished when all three triggering distances are calculated and recorded, otherwise, continuing detection.

S904, executing a characteristic standard deviation identification algorithm according to the collected multi-record data, specifically steps S300-S500.

S905, when the identification fails, an error is reported and characteristic data (namely collected multi-record data) is uploaded, when the identification succeeds, the control module controls the second motor to change the steering direction of the guide structure (specifically the guide groove), and then the control module controls the first motor to control the tray (carrying tray) to rotate, so that the articles can fall into the storage cabinet through the second through hole and enter the corresponding storage space under the buffering and guiding effects of the guide structure.

And S906, generating and displaying the two-dimensional code, entering a user collection stage, determining the payment state of the user, ending the user collection stage if the payment is successful, returning the garbage classification device to the standby state, recording abnormal data if the payment state is overtime, namely the payment state is abnormal, uploading the abnormal data to a server, and returning the garbage classification device to the standby state.

The embodiment of the invention has the following advantages; 1. firstly, the sorting device has enough arrangement quantity, which is convenient for being searched and used, secondly, the sorting and recycling are more convenient and executed, for example, only certain garbage (such as bottle-shaped recycled products, including pop-top cans, mineral water bottles, beer glass bottles and the like) are paid and recycled, and the enthusiasm of people for garbage collection and recycling work is improved. 2. The method changes the conventional mode of deploying a large amount of fixed recovery equipment into a new mode of reserving individual positions to deploy fixed recovery points and increasing unmanned trolley carrying equipment in a park. Thereby satisfying the service requirements of the recycled goods in a specific scene (such as a park, a business district, a dormitory, a residential district, etc.). 3. The multi-sensor multi-data fusion replaces a visual identification mode, reliable article identification and classification functions can be completed by combining corresponding algorithms, and meanwhile, the energy consumption and equipment cost of a single product are greatly reduced. 4. The multi-sensor and multi-data fusion mode has good effect of identifying the recycled articles, does not depend on specific bar codes but directly identifies and classifies according to the material of the articles, thereby omitting traceability measures and monitoring camera equipment, and reducing operation and maintenance cost and power consumption.

The embodiment of the invention also provides another garbage classification device based on the unmanned trolley, which comprises a processor and a memory;

the memory is used for storing programs;

the processor is used for executing programs to realize the classification method of the embodiment of the invention. The device of the embodiment of the invention can realize the function of classification. The device can be any intelligent terminal such as a mobile phone, a tablet computer, a Personal Digital assistant (PDA for short), a Point of Sales (POS for short), and the like.

The contents in the above method embodiments are all applicable to the present apparatus embodiment, the functions specifically implemented by the present apparatus embodiment are the same as those in the above method embodiments, and the advantageous effects achieved by the present apparatus embodiment are also the same as those achieved by the above method embodiments.

The embodiment of the present invention further provides a computer-readable storage medium, where a program is stored in the computer-readable storage medium, and the program is executed by a processor to implement the classification method according to the foregoing embodiment of the present invention.

Embodiments of the present invention also provide a computer program product comprising instructions, which when run on a computer, cause the computer to perform the classification method of the aforementioned embodiments of the invention.

The terms "first," "second," "third," "fourth," and the like in the description of the application and the above-described figures, if any, are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used is interchangeable under appropriate circumstances such that the embodiments of the application described herein are capable of operation in sequences other than those illustrated or described herein. Furthermore, the terms "comprises," "comprising," and "having," and any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed, but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus.

It should be understood that in the present application, "at least one" means one or more, "a plurality" means two or more. "and/or" for describing an association relationship of associated objects, indicating that there may be three relationships, e.g., "a and/or B" may indicate: only A, only B and both A and B are present, wherein A and B may be singular or plural. The character "/" generally indicates that the former and latter associated objects are in an "or" relationship. "at least one of the following" or similar expressions refer to any combination of these items, including any combination of single item(s) or plural items. For example, at least one (one) of a, b, or c, may represent: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", wherein a, b, c may be single or plural.

In the several embodiments provided in the present application, it should be understood that the disclosed apparatus and method may be implemented in other ways. For example, the above-described apparatus embodiments are merely illustrative, and for example, a division of a unit is merely a logical division, and an actual implementation may have another division, for example, a plurality of units or components may be combined or integrated into another system, or some features may be omitted, or not executed. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be an indirect coupling or communication connection through some interfaces, devices or units, and may be in an electrical, mechanical or other form. Units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of the embodiment. In addition, functional units in the embodiments of the present application may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit. The integrated unit can be realized in a form of hardware, and can also be realized in a form of a software functional unit.

The integrated unit, if implemented in the form of a software functional unit and sold or used as a stand-alone product, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present application may be substantially implemented or contributed to by the prior art, or all or part of the technical solution may be embodied in a software product, which is stored in a storage medium and includes multiple instructions for causing a computer device (which may be a personal computer, a server, or a network device) to execute all or part of the steps of the method of the embodiments of the present application. And the aforementioned storage medium includes: various media capable of storing programs, such as a usb disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk, or an optical disk.

The above embodiments are only used to illustrate the technical solutions of the present application, and not to limit the same; although the present application has been described in detail with reference to the foregoing embodiments, it should be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions in the embodiments of the present application.