Autonomous sterilization and disinfection device in inpatient ward scene

Abstract

The invention designs an autonomous sterilization and disinfection device in an inpatient ward scene. The autonomous sterilization and disinfection device comprises a sterilization and disinfection vehicle body, the control module, the sensing module, the movement module and the disinfection module are all mounted on the disinfection and sterilization vehicle body; according to the unmanned sterilization vehicle, autonomous routing inspection can be achieved in the sickroom scene of the inpatient department, disinfection is conducted according to environment information, the unmanned sterilization vehicle is provided with a depth camera and a laser radar, the environment information can be sensed in real time, and therefore the autonomous path planning and navigation obstacle avoidance functions in multiple sickrooms are achieved; in the navigation process, the unmanned sterilization vehicle detects information of personnel and sterilized articles in real time through the depth camera, and then controls the sterilization module to realize a graded sterilization function in different environments through a self-designed graded sterilization control algorithm, and sprays different amounts of disinfectant, so that intelligent sterilization of the environment of the inpatient department is realized; the disinfection efficiency is improved, the disinfection cost is saved, and the working intensity and the infection risk of epidemic prevention personnel are effectively reduced.

Description

Technical Field

[0001] The present invention belongs to the technical field of medical disinfection vehicles, and particularly relates to an autonomous sterilization and disinfection device for the inpatient ward scenario. Background Art

[0002] Disinfection work is a crucial part of the daily operation of medical institutions and is crucial for ensuring the environmental health and safety of medical facilities. In recent years, large-scale cross-infections between patients and medical staff have occurred in hospitals, which not only causes a reduction in the medical staff but also highlights the urgency of ensuring the environmental health of hospitals.

[0003] The problems existing in the current technology are as follows: The disinfection vehicles currently in use in hospitals cannot achieve functions such as fully autonomous navigation disinfection and automatic obstacle avoidance, and autonomous navigation disinfection vehicles do not involve the use scenarios of inpatient departments. Moreover, the spraying amount of disinfectant during the operation of the current autonomous navigation disinfection vehicle is a fixed value and cannot control the spraying amount of disinfectant according to the specific environment, resulting in inefficient utilization of disinfectant.

[0004] The above information disclosed in the background art section is only used to enhance the understanding of the background of the present disclosure, and thus it may include information that does not constitute the prior art known to those of ordinary skill in the art. Summary of the Invention

[0005] Aiming at the problems existing in the prior art, the present invention aims to provide an autonomous sterilization and disinfection device for the inpatient ward scenario to solve the above problems existing in the prior art.

[0006] To solve the above technical problems, the present invention adopts the following technical solutions: An autonomous sterilization and disinfection device for the inpatient ward scenario of the present invention includes: a disinfection and sterilization vehicle body; a control module, a sensing module, a motion module, and a disinfection module, all of which are installed on the disinfection and sterilization vehicle body; encoder DC reduction motors are installed at the four corners of the bottom of the disinfection and sterilization vehicle body and are equipped with Mecanum wheels; a lithium battery pack is fixedly connected to one side inside the disinfection and sterilization vehicle body to directly supply power to the STM32 control board above the top plate, and a Jetson Nano development board is fixedly connected to the other side, and the STM32 control board supplies power to the Jetson Nano development board; a depth camera is fixedly installed at the front end of the top plate and is connected to the Jetson Nano development board, an STM32 control board and a lidar are installed in the middle, and a disinfection water tank and an ultrasonic atomization sheet control board are fixedly installed at the rear. The STM32 control board is installed in the middle of the top plate and is connected to the encoder DC reduction motor of the chassis. A lidar is fixedly installed above the STM32 control board and is connected to the internal Jetson Nano development board. A disinfection water tank and an ultrasonic atomization sheet control board are fixedly installed at the rear of the top plate, and the ultrasonic atomization sheet control board is connected to the internal Jetson Nano development board.

[0007] In a preferred embodiment, the control module includes a Jetson Nano development board and an STM32 control board, which are connected by a serial communication line. The STM32 control board supplies power to the Jetson Nano development board and conducts data transmission through the serial communication line to achieve mutual data transmission.

[0008] In a preferred embodiment, the sensing module includes a depth camera and a lidar, which are connected to the Jetson Nano development board through a serial communication line and work simultaneously to synchronously collect environmental information for processing.

[0009] In a preferred embodiment, the motion module includes four encoder DC reduction motors of the chassis and supporting Mecanum wheels, which adopt a pendulum suspension system and are connected to the STM32 control board as a control source to control the Mecanum wheels to move, achieving movement at any angle, thereby completing flexible obstacle avoidance of obstacles.

[0010] In a preferred embodiment, the disinfection module includes an ultrasonic atomization sheet control board, a disinfection water tank, and ultrasonic atomization sheets. There are three circular holes on one side of the water tank for installing cylindrical water guiding cotton strips, and a water filling port for liquid replenishment on the other side of the water tank. The ultrasonic atomization sheet control board is connected with three ultrasonic atomization sheets, which are installed above the circular holes. During operation, ultrasonic waves are emitted by vibrating the piezoelectric ceramic transducer (atomization sheet) in the liquid, and the liquid is finally torn into tiny droplets at the interface to form atomization, thereby realizing the spraying of the disinfectant.

[0011] A hierarchical disinfection control algorithm for an autonomous sterilization and disinfection device in the scenario of an inpatient ward. The hierarchical disinfection control algorithm includes the following steps: S1. Collect various videos or pictures that need to be disinfected in the hospital and preprocess the videos or pictures; S2. Make a data set of the target object; S3. Train a YOLO neural network model to extract the features of various objects; S4. Collect hospital environment images through a depth camera and input them into the trained YOLO model; S5. Extract the categories and confidence levels of the objects recognized by YOLO and input them into a mathematical formula to calculate the disinfection amount S; S6. Identify which disinfection interval this disinfection amount belongs to, and transmit the signal to the ultrasonic atomization sheet control board to control the atomization sheet generator to achieve hierarchical disinfection.

[0012] In a preferred embodiment, training the YOLO neural network model in S3 includes the following steps: S31. Establish an algorithm model and set model parameters; S32. Obtain the total number of steps of the model and the resolution of the model input image; S33. Statistically analyze all samples and perform visualization to obtain the number of iterations of the preliminary training, the mAP value and Results value of the initialized model; S34. Repeatedly train the model, record the training results and data of each model until the data deviation value meets the expected value.

[0013] In a preferred embodiment, the mathematical formula for the disinfection amount S is , this formula calculates the disinfection supply amount S; A i is the disinfection weight of the i-th type of object; p i is the confidence of the detection target; x max , x min , y max , y min are the relative position parameters of the bounding box; B is the total number of bounding boxes in the image; C is the number of categories of detection objects in the image.

[0014] The mathematical formula for the disinfection amount S, the disinfection weight A of the i-th type of object i The obtaining of, includes the following steps: S1. Start the autonomous sterilization device to complete a working cycle of the autonomous sterilization device; S2. Statistically analyze by computer the number of times various disinfection objects are recognized by the autonomous sterilization device within a working cycle; S3. Sort in ascending order according to the number of times the disinfection object appears within a working cycle, and the corresponding serial number is i, i = 1, 2, 3,...; S4. According to the sorting result, the disinfection weights of various disinfection objects are 5i + 1 in sequence, that is, 5, 10, 15,...

[0015] Compared with the prior art, the present invention has the following beneficial technical effects: The autonomous sterilization and disinfection device of the present invention, when used in the inpatient department scenario, can mark the location of each ward, and then autonomously traverse the wards, using a depth camera and a lidar for obstacle avoidance during autonomous navigation to automatically avoid people and obstacles; at the same time, use a depth camera to collect environmental information, identify the disinfection objects in the picture using the YOLO algorithm, and then through the self-designed hierarchical disinfection control algorithm, achieve hierarchical disinfection in different disinfection environments, that is, this autonomous sterilization and disinfection device can perform autonomous sterilization and disinfection on the inpatient department ward environment, and spray different amounts of disinfectant according to different environments, solving the problems mentioned in the background technology. Brief Description of the Drawings

[0016] To more clearly illustrate the technical solutions in the embodiments of the present invention, the following will briefly introduce the accompanying drawings required for the description of the embodiments. Obviously, the accompanying drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other accompanying drawings can be obtained based on these drawings.

[0017] Figure 1 Schematic diagram of the overall structure of the autonomous sterilization and disinfection device provided by the present invention; Figure 2 Schematic diagram of the disinfection water tank structure of the autonomous sterilization and disinfection device provided by the present invention; Figure 3 Schematic diagram of the chassis structure of the autonomous sterilization and disinfection device provided by the present invention Figure 4 Schematic diagram of the control panel of the autonomous sterilization and disinfection device provided by the present invention; Figure 5 Schematic diagram of the operation process of the autonomous sterilization and disinfection device provided by the present invention; Explanation of reference numerals: 1, main body of the disinfection and sterilization vehicle; 2, Jetson Nano development board; 3, STM32 control board; 4, control board for ultrasonic atomization sheet; 5, depth camera; 6, lidar; 7, disinfection water tank; 8, ultrasonic atomization sheet; 9, lithium battery pack; 10, encoder DC reduction motor; 11, Mecanum wheel; 12, circular hole; 13, water injection port. Specific implementation manners

[0018] In order to make the objectives, technical solutions and advantages of the present invention clearer, the following will further describe the present invention in detail with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, rather than all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts fall within the protection scope of the present invention.

[0019] Embodiment 1 Refer to the specification appendix Figure 1, an autonomous sterilization and disinfection device in the inpatient ward scenario of the present invention, comprising: a disinfection sterilization vehicle body 1; a control module, a sensing module, a motion module, and a disinfection module, all installed on the disinfection sterilization vehicle body 1; encoder DC reduction motors 10 are installed at the four corners of the bottom of the disinfection sterilization vehicle body 1 and are equipped with Mecanum wheels 11; a lithium battery pack 9 is fixedly connected to one side inside the disinfection sterilization vehicle body 1, directly supplying power to the STM32 control board 3 above the top plate, and a Jetson Nano development board 2 is fixedly connected to the other side, and the STM32 control board 3 supplies power to the Jetson Nano development board 2; a depth camera 5 is fixedly installed at the front end of the top plate, connected to the Jetson Nano development board 2, an STM32 control board 3 and a lidar 6 are installed in the middle, and a disinfection water tank 7 and an ultrasonic atomization sheet control board 4 are fixedly connected to the rear. The control module includes a Jetson Nano development board 2 and an STM32 control board 3, which are connected by a serial communication line and a power line, and the STM32 control board 3 supplies power to the Jetson Nano development board 2. The sensing module includes a depth camera 5 and a lidar 6, which are connected to the Jetson Nano development board 2 through a serial communication line. The motion module includes four encoder DC reduction motors 10 and supporting Mecanum wheels 11 at the rear of the chassis, with the STM32 control board 3 as the control source. The disinfection module includes an ultrasonic atomization sheet control board 4, a disinfection water tank 7 and an ultrasonic atomization sheet 8. There are three circular holes 12 on one side of the water tank for installing cylindrical water guiding cotton strips, and a water filling port 13 for liquid replenishment on the other side. The Mecanum wheel 11 adopts a pendulum suspension system.

[0020] A hierarchical disinfection control algorithm for an autonomous sterilization and disinfection device in the inpatient ward scenario, the hierarchical disinfection control algorithm comprising the following steps: S1. Collect various videos or pictures that need to be disinfected in the hospital and preprocess the videos or pictures; S2. Make a data set of target objects; S3. Train a YOLO neural network model to extract the features of various objects; S4. Collect hospital environment images through the depth camera 5 and input them into the trained YOLO model; S5. Extract the categories and confidence levels of the objects recognized by YOLO and input them into a mathematical formula to calculate the disinfection amount S; S6. Identify which disinfection interval the disinfection amount belongs to, and transmit the signal to the ultrasonic atomization sheet control board 4, thereby controlling the ultrasonic atomization sheet 8 to achieve hierarchical disinfection.

[0021] In S1, the collected videos and pictures are processed by manual system annotation.

[0022] The dataset object of S2 includes people, inpatient beds, medical waste bins, medical wheelchairs, benches, and bedside cabinets, and the pictures are labeled and converted to generate label files.

[0023] Training the YOLO neural network model in S3 includes the following steps: S31. Establish an algorithm model and set model parameters; S32. Obtain the total number of model steps and the resolution of the model input picture; S33. Statistically analyze all samples and visualize them to obtain the number of iterations of the preliminary training, the mAP value and the Results value of the initialized model; S34. Repeatedly train the model, record the training results and data of each model until the data deviation value meets the expected value.

[0024] The mathematical formula for the disinfection quantity S is , and this formula calculates the disinfection supply quantity S; A i is the disinfection weight of the i-th type of object; p i is the confidence of the detection target; x max 、x min 、y max 、y min are the relative position parameters of the bounding box; B is the total number of bounding boxes in the image; C is the number of categories of detection objects in the image.

[0025] The mathematical formula for the disinfection quantity S, the disinfection weight A of the i-th type of object i The obtaining of, includes the following steps: S1. Start the autonomous sterilization device to complete a working cycle of the autonomous sterilization device; S2. Statistically analyze the number of times the autonomous sterilization device recognizes various disinfection objects within a working cycle through a computer; S3. Sort the number of times the disinfection objects appear within a working cycle from small to large, and the corresponding serial numbers are i, i = 1, 2, 3,...; S4. According to the sorting results, the disinfection weights of various disinfection objects are 5i + 1 in sequence, that is, 5, 10, 15,...

[0026] In this example, the specific working process is as follows: Refer to the attached instructions Figure 2 , add the disinfectant to the disinfection water tank through the water injection port, and turn on the switch of the ultrasonic atomization sheet control board. By default, one atomization sheet starts to work for the lowest level of disinfection.

[0027] Refer to the attached instructions Figure 5, when the device is used in a new scenario for the first time, it is necessary to control the autonomous disinfection and sterilization vehicle to turn on the radar and cruise around the working floors of the inpatient department to complete the drawing of the two-dimensional radar map of the scene environment, and mark each ward on the map with serial numbers 1, 2, 3, and so on.

[0028] Refer to the appendix of the instruction manual Figure 5 , start the autonomous sterilization and disinfection device. The vehicle starts from the starting point and, according to the constructed radar map, automatically plans a route to the ward numbered 1 and navigates there. After arrival, it successively plans routes to the wards with subsequent serial numbers and navigates to them.

[0029] Refer to the appendix of the instruction manual Figure 4 , during operation, the depth camera reads the environmental information in front of the vehicle, transmits the image information to the Jetson Nano development board, which is analyzed by the YOLO algorithm configured in the Jetson Nano development board. Then, the self-designed hierarchical disinfection control algorithm obtains the disinfection amount S, and the specific disinfection level is obtained from the hierarchical disinfection table. The control signal is transmitted to the ultrasonic atomization sheet control board.

[0030] Refer to the appendix of the instruction manual Figure 4 , after the control signal is transmitted to the ultrasonic atomization sheet control board, the ultrasonic atomization sheet control board controls the number of working ultrasonic atomization sheets to be turned on. If it is in the first-level disinfection state, the middle ultrasonic atomization sheet is turned on; if it is in the second-level disinfection state, the ultrasonic atomization sheets on both sides are turned on; if it is in the third-level disinfection state, all ultrasonic atomization sheets are turned on.

[0031] After reaching the last marked ward, the autonomous sterilization device automatically plans the path to return to the starting point from the current position, and at this time, one cruise disinfection work process ends.

[0032] During the navigation of the autonomous sterilization and disinfection device, if a pedestrian suddenly appears on the planned driving path, the sterilization vehicle activates the autonomous obstacle avoidance function. The vehicle retreats a certain distance, scans the radar again to plan a path, bypasses the pedestrian, and continues to complete the operation.

[0033] The above is only the preferred implementation mode of the present invention. It should be understood that the present invention is not limited to the form disclosed herein, should not be regarded as excluding other embodiments, but can be used in various other combinations, modifications, and environments, and can be changed within the scope of the concept described herein through the above teachings or the technology or knowledge in related fields. And the changes and alterations made by those skilled in the art that do not depart from the spirit and scope of the present invention should all be within the protection scope of the appended claims of the present invention.

Claims

1. An autonomous sterilization and disinfection device for an inpatient ward, characterized in that: include: A disinfection and sterilization vehicle body (1); a control module, a sensing module, a motion module, and a disinfection module are all installed on the disinfection and sterilization vehicle body (1); encoder DC reduction motors (10) are installed at the four corners of the bottom of the disinfection and sterilization vehicle body (1) and are matched with Mecanum wheels (11); a lithium battery pack (9) is fixedly connected to one side of the disinfection and sterilization vehicle body (1) to directly supply power to an STM32 control board (3) above the top plate, and a Jetson Nano development board (2) is fixedly connected to the other side of the disinfection and sterilization vehicle body (1), and the STM32 control board (3) supplies power to the Jetson Nano development board (2); a depth camera (5) is fixedly installed at the front end of the top plate and is connected to the Jetson Nano development board (2); an STM32 control board (3) and a laser radar (6) are installed in the middle; and a disinfection water tank (7) and an ultrasonic atomization sheet control board (4) are fixed at the rear.

2. The autonomous sterilization and disinfection device for an inpatient ward according to claim 1 is characterized in that: The control module comprises a Jetson Nano development board (2) and an STM32 control board (3), which are connected via a serial communication line and a power line, and the STM32 control board (3) supplies power to the Jetson Nano development board (2).

3. The autonomous sterilization and disinfection device for an inpatient ward according to claim 1 is characterized in that: The perception module comprises a depth camera (5) and a laser radar (6), and is connected to a Jetson Nano development board (2) via a serial communication line.

4. The autonomous sterilization and disinfection device for an inpatient ward according to claim 1 is characterized in that: The motion module comprises four encoder DC reduction motors (10) and matching Mecanum wheels (11) at the rear of the chassis, and uses an STM32 control board (3) as a control source.

5. The autonomous sterilization and disinfection device for an inpatient ward according to claim 1 is characterized in that: The disinfection module comprises an ultrasonic atomizer control panel (4), a disinfection water tank (7) and an ultrasonic atomizer (8), one side of the water tank is provided with three circular holes (12) for installing cylindrical water-guiding cotton strips, and the other side is provided with a water injection port (13) for liquid replenishment.

6. The autonomous sterilization and disinfection device for an inpatient ward according to claim 4 is characterized in that: The Mecanum wheel (11) adopts a pendulum suspension system.

7. According to claim 3, the hierarchical disinfection control algorithm of the autonomous sterilization and disinfection device in the inpatient ward scenario is characterized in that: The hierarchical disinfection control algorithm comprises the following steps: S1. Collect all kinds of videos or pictures that need to be disinfected in the hospital and pre-process them; S2, create a dataset of the target object; S3, train the YOLO neural network model to extract the features of each category of objects; S4, collecting hospital environment images through the depth camera (5) and inputting the trained YOLO model; S5. Extract the category and confidence of the object recognized by YOLO, and input them into the mathematical formula to calculate the disinfection amount S; S6, identifying which disinfection interval the disinfection amount belongs to, transmitting a signal to the ultrasonic atomization plate control board (4), thereby controlling the ultrasonic atomization plate (8) to achieve graded disinfection.

8. The hierarchical disinfection control algorithm of the autonomous sterilization and disinfection device in the inpatient ward scenario according to claim 7 is characterized in that: In S1, the collected videos and pictures are manually annotated by the system; the objects in the dataset of S2 include people, inpatient beds, medical waste bins, medical wheelchairs, benches, bedside tables, and the pictures are annotated and converted to generate label files; Training the YOLO neural network model in S3 includes the following steps: S31, establish an algorithm model and set model parameters; S32, obtaining the total step length of the model and the resolution of the model input image; S33, count and visualize all samples to obtain the number of iterations of the initial training, the mAP value of the initialized model, and the Results value; S34. Repeatedly train the models and record the training results and data of each model until the data deviation value meets the expected value.

9. The hierarchical disinfection control algorithm of the autonomous sterilization and disinfection device in the inpatient ward scenario according to claim 7 is characterized in that: The mathematical formula for the disinfection volume S is: , this formula calculates the consumption S; A i is the disinfection weight of the i-th type of object; p i is the confidence of the detected target; x max 、x min ,y max ,y min is the relative position parameter of the bounding box; B is the total number of bounding boxes in the image; C is the number of categories of detected objects in the image.

10. The hierarchical disinfection control algorithm of the autonomous sterilization and disinfection device in the inpatient ward scenario according to claim 9 is characterized in that: Mathematical formula of disinfection quantity S, disinfection weight A of the i-th object i The obtaining of includes the following steps: S1. Start the autonomous sterilization device to complete a working cycle; S2. Count the number of times the autonomous sterilization device identifies various types of disinfection objects within a working cycle by computer; S3, sorting the disinfection objects according to the number of times they appear in a working cycle from small to large, with the corresponding sequence number being i, i=1, 2, 3, ...; S4. According to the ranking results, the disinfection weights of various disinfection objects are 5i, i.e. 5, 10, 15, ...

Images