Robot microbiological detection preparation platform

Abstract

The invention discloses a robot microbiological detection preparation platform, and relates to the technical field of microbiological detection and pretreatment, and the system comprises the following components: a microbiological intelligent detection module which is based on a deep learning algorithm, enables a robot to automatically learn and adjust a strategy according to environment feedback, and adopts a behavior gradient Q learning algorithm and an XCSG integration algorithm to obtain a microbiological detection result; according to the system and the method, multiple types, small batches and multiple batches of microbiological detection operations are realized, the processes of sample homogenization, culture medium preparation, microbiological incubation, scribing, inoculation and detection are automatically completed, real-time monitoring and data analysis are realized, and an unmanned intelligent operation detection process is realized. Automation and intelligentization of microorganism detection are achieved, the system can automatically complete the whole process from sample homogenization and culture medium preparation to microorganism incubation, lineation, inoculation and detection, interference of human factors is reduced, and the reliability of detection results is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of microbial detection and pre-treatment, and in particular to a robot microbial detection and preparation platform. Background Art

[0002] With the continuous advancement of science and technology and the development of globalization, the importance of microbial testing in various fields has become increasingly prominent. From medical diagnosis to food safety, to environmental monitoring, microbial testing plays a vital role. Especially in the field of biomedicine, microbial testing is an important indicator to measure the sanitary quality of products and plays a decisive role in ensuring the safety of products. In addition, as people's attention to health and safety continues to increase, the requirements for microbial testing technology are becoming increasingly stringent. Although traditional microbial testing methods have met the needs to a certain extent, they are limited by manual operation, detection efficiency and accuracy, and can no longer meet the rapid development needs of modern society.

[0003] At present, traditional microbial detection technology mainly relies on manual operation, which is not only inefficient but also easily affected by human factors, resulting in unstable test results. In addition, traditional methods usually require a long detection cycle and cannot meet the needs of rapid diagnosis. In the medical field, this may lead to delayed treatment of diseases and even more serious consequences. In the field of food safety, a long detection cycle may also cause unqualified products to enter the market and endanger public health. At the same time, traditional microbial detection methods also have shortcomings in data processing and analysis, and cannot make full use of modern information technology to improve detection efficiency and accuracy.

[0004] However, traditional microbial detection technology is also vulnerable to human factors, such as operational errors and subjective judgment, because the detection process mostly relies on manual operation, resulting in unstable test results. It has limited detection capabilities for new or mutant microorganisms and is difficult to adapt to changing detection needs. In addition, traditional methods usually require expensive equipment and reagents, which increases detection costs, and have difficulties in processing large amounts of data, making it difficult to achieve real-time updating and sharing of data, limiting the further development of detection technology. Therefore, the research and development and application of microbial intelligent detection and pretreatment based on robots and big data are of great significance for solving the shortcomings and disadvantages of traditional technologies. Summary of the invention

[0005] The purpose of the present invention is to make up for the shortcomings of the existing technology and provide a robotic microbial detection and preparation platform, which can use big data technology to comprehensively analyze and process the detection data, monitor the detection process in real time through the random forest algorithm model, effectively warn of potential quality risks, and ensure the reliability and accuracy of the detection results. This process not only speeds up the output of detection results, but also improves the scientific nature of detection and the accuracy of decision-making through intelligent analysis, providing more powerful detection support for industries such as medical care, food, and environmental protection.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions: a robot microorganism detection and preparation platform, characterized in that the system includes the following components:

[0007] Microbial intelligent detection module: Based on deep learning algorithms, the robot can autonomously learn and adjust strategies according to environmental feedback. It uses behavioral gradient Q learning algorithms and XCSG integrated algorithms to achieve multi-type, small-batch, and multi-batch microbial detection operations, automatically complete sample homogenization, culture medium preparation, microbial incubation, streaking, inoculation, and detection processes, and conduct real-time monitoring and data analysis to achieve unmanned intelligent operation and detection processes.

[0008] Microbial culture module: provides the environmental conditions required for microbial growth, including temperature, humidity, and light, to ensure the normal growth and reproduction of microorganisms and provide reliable samples for subsequent detection and analysis;

[0009] Microbial intelligent analysis module: It uses multi-degree-of-freedom robots, servo slides, and flexible grippers, in conjunction with IoT technology, and works with water baths and centrifuges to automatically detect and analyze microbial samples, obtain information on the types and quantities of microorganisms, and achieve full automation of the entire chain from sample processing and detection to result analysis;

[0010] Data processing module: Analyze and process the test data using big data technology, and use the random forest algorithm model to intelligently analyze the data during the test process, monitor the status of robots and test equipment in real time, and promptly warn of potential quality risks to ensure the accuracy and reliability of the test results;

[0011] IoT technology integration module: Integrates RFID, WSN and visual recognition technologies to achieve comprehensive digital management of the testing process, including real-time monitoring and data transmission of reagent types, sample data, and equipment operating status, forming an unmanned microbial intelligent testing laboratory to achieve safety and efficiency of testing;

[0012] Automatic sterilization module: It uses efficient and safe ultraviolet and ozone sterilization technology, and cooperates with the microbial detection robot to automatically sterilize the detected microbial samples, instruments and equipment, and laboratory environment to ensure that no secondary contamination is generated during the detection process and to ensure a sterile environment for the detection process.

[0013] Furthermore, the microbial intelligent detection module realizes autonomous learning of the robot based on a deep learning algorithm. By taking various tentative actions on the perceived environmental state, it obtains the reward and penalty signals of the environmental state for such tentative actions, and continuously adjusts the learning strategy to obtain larger rewards and smaller penalties, which are fed back to the robot to realize rapid convergence of its learning, thereby achieving the purpose of autonomous learning and flexible adaptation to multiple types of detection operations.

[0014] Furthermore, the microbial intelligent detection module learns to perform optimal microbial detection operations by interacting with the environment through a behavioral gradient Q learning algorithm, and trains the robot to select detection strategies for culture medium preparation and microbial incubation conditions based on environmental feedback of microbial sample characteristics and detection equipment status. The behavioral gradient method improves the strategy by directly optimizing the strategy parameters, while Q learning learns an action value function to indicate the expected return after taking an action in a given state, allowing for more direct optimization of strategies and adaptation to continuous action spaces. The Q learning update formula is: Among them, Q(s t ,a t ) is in state s at time step t t Take action a t The expected return is α, the learning rate is r t+1 Is to take action a t The reward obtained, γ is the discount factor, which indicates the importance of future rewards relative to immediate rewards, is in the next state s t+1 The robot continuously interacts with the environment and updates the Q table, thereby gradually learning the strategy of choosing the best action under different states.

[0015] Furthermore, the microbial intelligent detection module will adopt the behavioral gradient Q learning algorithm and the XCSG integration algorithm. By combining the robot learning planning rules with the model function under the environmental model, the continuous state and continuous action are integrated with the environment and result feedback to realize sequential decision-making, and the weight matrix value is mapped to XCSG, and the robot learning convergence speed is increased exponentially from the result reward strategy, so as to achieve the purpose of strengthening the robot's autonomous learning to cope with the goal of flexible robot adaptation in different experiments and small batch and multi-batch experiments.

[0016] Furthermore, the data processing module uses the standardization and mean algorithm in big data technology to analyze and process the microbial detection data, and the standardization and mean algorithm formula is: x is the raw microbial data, which integrates data from multiple testing points, multiple batches, and multiple types for efficient storage, query, and analysis. It collects data information on temperature, humidity, microbial types, and quantity during the microbial testing process, monitors the status of robots and testing equipment in real time, promptly identifies potential quality risks, and evaluates the accuracy and reliability of test results.

[0017] Furthermore, the data processing module uses a data set processed by big data technology to construct a random forest model, selects feature data from the test data, including key variables such as sample concentration, incubation time, and equipment temperature, as the basis for constructing each tree. For the construction of each decision tree, the algorithm will find the best split point in the selected feature subset to divide the data. This process is repeated so that each tree independently classifies and regresses the samples. The algorithm formula is: T(x) = argmax c∈C ∑ t∈forest I(t(x)=c), T(x) represents the predicted category of the input sample x, C is the set of all categories, t(x) represents the predicted category of the tth tree for the sample x, I is the indicator function, forest represents the entire random forest, and each decision tree is trained through training data. Each decision tree will divide the data according to the selected features to maximize the accuracy of classification and minimize the error of regression. The test data set is used to evaluate the performance of the model, and the trained random forest model is used to predict new data. The predicted results are compared with the actual results, and the accuracy, precision and recall of the model are evaluated. The prediction results are output, and the model is optimized and adjusted according to the evaluation indicators. The prediction results of the random forest model are expressed as the average and voting results of the prediction results of multiple decision trees, that is: F(x) is the final prediction result of the random forest model for sample x, and M is the number of decision trees in the random forest.

[0018] Furthermore, the radio frequency identification (RFID) technology in the IoT technology integration module is used to track and manage the reagents and sample items required for microbial detection to ensure their accuracy and traceability. The wireless sensor network (WSN) is used to monitor the environmental conditions of temperature, humidity, and light in the laboratory, as well as the operating status of the detection equipment. By collecting and analyzing these data in real time, it is ensured that the microorganisms are in the best growth and detection environment. Visual recognition technology is used to automatically identify and classify microbial samples.

[0019] Furthermore, the IoT technology integration module integrates and processes multi-source data from RFID, WSN and visual recognition through Kalman filtering in the data fusion algorithm, eliminates redundant information, extracts useful microbial data, improves data integrity and accuracy, and recursively combines prior knowledge and observation data to update the estimate of the system state. The algorithm formula of the Kalman filter prediction and update steps is: Prediction: renew: in, represents state estimation, F is the state transfer matrix, B is the control input influence matrix, u is the control input, K is the Kalman gain, z is the measurement value, and H is the observation matrix. Different multi-source data are integrated, including the microbial detection identity and location information provided by RFID, the environmental parameters transmitted by WSN, and the image and video information captured by the visual recognition system. Intelligent fusion improves data quality and provides a more comprehensive and accurate information basis for decision-making.

[0020] Furthermore, the ultraviolet sterilization technology in the automatic sterilization module destroys the DNA structure of microorganisms, uses ultraviolet lamps to generate ultraviolet radiation, and irradiates the sterilized microbial samples, instruments and equipment and the environment, thereby preventing their replication and reproduction, thereby achieving the sterility of the equipment. The ozone sterilization technology reacts with the cell walls, cell membranes and enzymes in the cells of microorganisms to destroy their structures, and generates ozone through an ozone generator, which is transported to the area that needs to be sterilized, thereby achieving equipment sterility.

[0021] Furthermore, the automatic sterilization module automatically adjusts the sterilization parameters through the PID control algorithm to ensure that the temperature, humidity, ultraviolet intensity and ozone concentration are maintained within the optimal range to achieve the best sterilization effect. The PID controller continuously compares the expected environmental parameters with the actual measured values, calculates the control signal, and then adjusts the output of the sterilization equipment, including changing the power of the ultraviolet lamp and controlling the output of the ozone generator to correct the deviation and maintain the stability of the system. The algorithm formula is: K p , K i and K d are the proportional coefficient, integral coefficient and differential coefficient respectively, represents the integral of the error over time, is the instantaneous rate of change of the error, that is, the differential of the error, and u(t) is the power and concentration control instructions adjusted at time t.

[0022] Compared with the existing technology, this intelligent microbial detection and sterilization system has the following beneficial effects:

[0023] 1. The robot detection system of the present invention realizes the automation and intelligence of microbial detection by introducing robots and deep learning algorithms, which not only greatly reduces the difficulty and error rate of manual operation, but also significantly improves the efficiency and accuracy of detection and sterilization. The use of robots enables the system to automatically complete the entire process from sample homogenization, culture medium preparation to microbial incubation, streaking, inoculation and detection, thereby reducing the interference of human factors and improving the reliability of detection results.

[0024] 2. The system of the present invention utilizes big data technology to process and analyze test data in real time, and can quickly and accurately obtain test results. Through the random forest algorithm model, the system can monitor the status of robots and testing equipment in real time, and promptly discover and warn of potential quality risks. This real-time and accuracy is crucial for microbial testing in the medical, food, and environmental protection fields, and can provide strong support for decision-making in related industries.

[0025] 3. The present invention introduces an automatic sterilization module and adopts efficient and safe sterilization technologies, such as ultraviolet sterilization and ozone sterilization, to ensure the sterility of the testing environment and effectively avoid secondary pollution that may occur during the testing process. At the same time, the implementation of the Internet of Things (IoT) technology integration module realizes the digital management of the testing process through RFID, WSN and visual recognition technology, enhances the safety and efficiency of the laboratory, and reduces the burden on the environment. This full-chain automation and intelligence not only improves the safety of testing work, but also reduces energy consumption through intelligent management, which meets the requirements of green and sustainable development.

[0026] Other advantages, objectives and features of the present invention will be set forth in part in the following description and, in part, will be apparent to those skilled in the art based on an examination of the following or may be taught from the practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention, and for ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0028] Figure 1 This is a process operation diagram of a robotic microbial detection and preparation platform.

[0029] Figure 2 Flowchart of a robotic microbial detection and preparation platform. DETAILED DESCRIPTION

[0030] The technical solutions in the embodiments of the present invention are described clearly and completely below. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0031] A robotic microbial detection and preparation platform, the system includes the following components:

[0032] Microbial intelligent detection module: Based on deep learning algorithms, the robot can autonomously learn and adjust strategies according to environmental feedback. It uses behavioral gradient Q learning algorithms and XCSG integrated algorithms to achieve multi-type, small-batch, and multi-batch microbial detection operations, automatically complete sample homogenization, culture medium preparation, microbial incubation, streaking, inoculation, and detection processes, and conduct real-time monitoring and data analysis to achieve unmanned intelligent operation and detection processes.

[0033] Microbial culture module: provides the environmental conditions required for microbial growth, including temperature, humidity, and light, to ensure the normal growth and reproduction of microorganisms and provide reliable samples for subsequent detection and analysis;

[0034] Microbial intelligent analysis module: It uses multi-degree-of-freedom robots, servo slides, and flexible grippers, in conjunction with IoT technology, and works with water baths and centrifuges to automatically detect and analyze microbial samples, obtain information on the types and quantities of microorganisms, and achieve full automation of the entire chain from sample processing and detection to result analysis;

[0035] Data processing module: Analyze and process the test data using big data technology, and use the random forest algorithm model to intelligently analyze the data during the test process, monitor the status of robots and test equipment in real time, and promptly warn of potential quality risks to ensure the accuracy and reliability of the test results;

[0036] IoT technology integration module: Integrates RFID, WSN and visual recognition technologies to achieve comprehensive digital management of the testing process, including real-time monitoring and data transmission of reagent types, sample data, and equipment operating status, forming an unmanned microbial intelligent testing laboratory to achieve safety and efficiency of testing;

[0037] Automatic sterilization module: It uses efficient and safe ultraviolet and ozone sterilization technology, and cooperates with the microbial detection robot to automatically sterilize the detected microbial samples, instruments and equipment, and laboratory environment to ensure that no secondary contamination is generated during the detection process and to ensure a sterile environment for the detection process.

[0038] Embodiment 1

[0039] This embodiment introduces in detail the specific application of a robotic microbial detection and preparation platform, demonstrating the powerful functions and advantages of the system in the field of microbial detection. The system not only realizes the full process automation and intelligence of microbial detection, but also realizes the comprehensive digital management of the detection process through the Internet of Things (IoT) technology, thereby improving the safety and efficiency of detection. At the same time, the application of the automatic sterilization module also ensures a sterile environment for the detection process, avoiding the occurrence of secondary contamination.

[0040] In the specific implementation, the system uses a deep learning algorithm to take various tentative actions on the perceived environmental state, obtain the reward and penalty signals of the environmental state for such tentative actions, and continuously adjust the learning strategy to obtain larger rewards and smaller penalties, and feedback to the robot to achieve rapid convergence of its learning, so as to achieve the purpose of autonomous learning and flexible adaptation to multi-type detection operations, so that the robot can self-learn and adjust its strategy according to environmental feedback. The behavioral gradient Q learning algorithm will be used: By integrating the XCSG integrated algorithm, the robot learning planning rules are combined with the model function under the environmental model, and the continuous state and continuous action are combined with the environment and result feedback to realize sequential decision-making. The weight matrix value is mapped to XCSG, and the robot learning convergence speed is increased exponentially from the result reward strategy, so as to achieve the purpose of strengthening the robot's autonomous learning to cope with the goals of different experiments and small batch and multi-batch experiments that require the robot to adapt flexibly, including automatic sample homogenization, automated preparation of culture medium, precise incubation of microorganisms, streaking inoculation and subsequent detection processes. The whole process does not require human intervention, ensuring high efficiency and consistency of detection. Through learning, the robot can select the best detection strategy according to the characteristics of the microbial sample and the status of the equipment, such as selecting the most suitable culture conditions to achieve the best detection results.

[0041] The microbial culture module provides the environmental conditions required for the growth of microorganisms, including temperature, humidity, and light, to ensure the normal growth and reproduction of microorganisms and provide reliable samples for subsequent detection and analysis.

[0042] The intelligent microbial detection and analysis module utilizes multi-degree-of-freedom robots, servo slides, and flexible gripper devices, combined with Internet of Things technology, to achieve linkage with water baths and centrifuge equipment, completing the fully automated process from sample processing to test result analysis. Through these sophisticated automated equipment, accurate information on microbial species and quantities can be obtained, thereby improving detection accuracy and efficiency.

[0043] The data processing module uses big data technology to conduct in-depth analysis of the collected microbial detection data through standardization and mean algorithms: Process data to ensure its accuracy and reliability. Apply the random forest algorithm model to monitor the working status of robots and testing equipment in real time, discover and warn potential quality risks in a timely manner, and provide data support for decision-making.

[0044] The IoT technology integration module integrates RFID technology, wireless sensor network (WSN) and visual recognition technology to achieve comprehensive digital management of the detection process. Through these technologies, the system can monitor information such as reagent type, sample data, and equipment operating status in real time to ensure efficient and safe operation of the laboratory, while also facilitating data tracking and traceability.

[0045] In addition, the automatic sterilization module uses ultraviolet and ozone sterilization technology, and works in collaboration with the microbial detection robot to automatically sterilize microbial samples, experimental equipment and laboratory environment to ensure a sterile environment and avoid secondary contamination during the detection process. The module uses a PID control algorithm to dynamically adjust the sterilization parameters to ensure that the sterilization conditions are always in the best state.

[0046] Embodiment 2

[0047] A robotic microbial detection and preparation platform in this embodiment particularly emphasizes the use of multi-degree-of-freedom robots, servo slides and flexible gripper devices, combined with Internet of Things (IoT) technology, and collaborative work with laboratory equipment such as water baths and centrifuges to perform automated detection and analysis of microbial samples. It also combines a random forest model to monitor the system status in real time, promptly warn of potential quality problems, and ensure the accuracy and reliability of the test results.

[0048] First, the microbial inspection data provided by the microbial intelligent detection module and the microbial cultivation module are received, and a multi-degree-of-freedom robot, servo slide and flexible gripper device are used to automatically complete the reception and pretreatment of microbial samples, including sample homogenization, to ensure that the samples are uniform and consistent, which is convenient for subsequent testing. The system integrates with the Internet of Things (IoT) technology, links water baths and centrifuges, and performs automated testing of microbial samples. The robot places the processed samples under appropriate testing conditions according to a predetermined procedure, and performs microbial cultivation, streaking and inoculation. During the testing process, various data on microbial growth are collected in real time through high-precision sensors, including but not limited to temperature, humidity, and microbial growth. The collected data is quickly analyzed using an integrated algorithm to determine the type and quantity of microorganisms and generate a detailed test report. The entire process does not require human intervention, and the entire chain from sample processing to result analysis is fully automated.

[0049] Then, the microbial test data from various test sites, batches and types were integrated and normalized and averaged using algorithms. x is the original microbial data, which integrates data from multiple testing points, multiple batches, and multiple types for efficient storage, query, and analysis. It collects data information on temperature, humidity, microbial species, and quantity during the microbial testing process, monitors the status of robots and testing equipment in real time, promptly discovers potential quality risks, and evaluates the accuracy and reliability of test results. It uses big data technology to monitor the integrated data in real time, and uses the random forest algorithm model to select feature data from the test data, including key variables such as sample concentration, incubation time, and equipment temperature, as the basis for building each tree. For the construction of each decision tree, the algorithm will find the best split point to divide the data in the selected feature subset. This process is repeated so that each tree independently classifies and regresses the samples. The algorithm formula is: T(x) = argmax c∈C ∑ t∈forest I(t(x)=c), T(x) represents the predicted category of input sample x, C is the set of all categories, t(x) represents the predicted category of sample x by the tth tree, I is the indicator function, forest represents the entire random forest, and the model is trained and verified through the training data set, and the model parameters are continuously adjusted to improve the accuracy, precision and recall of the prediction. The model is applied to new data prediction and compared with the actual results to evaluate the model performance, and the model is adjusted according to the feedback of the evaluation results to continuously optimize the accuracy and reliability of the detection results. The prediction results of the random forest model are expressed as the average and voting results of the prediction results of multiple decision trees, that is: F(x) is the final prediction result of the random forest model for sample x, and M is the number of decision trees in the random forest. It performs intelligent analysis on the status of detection equipment and robots, timely detects potential quality risks and issues early warnings.

[0050] Finally, RFID technology is used to track and manage reagents and samples to ensure their correctness and full traceability. WSN technology monitors the laboratory's environmental conditions (such as temperature, humidity, and light) and equipment operating status to achieve real-time data collection. Visual recognition technology is used to automatically identify microbial samples, realize automatic classification of samples, and improve detection efficiency and accuracy. Kalman filtering and other data fusion algorithms are used to predict: renew: Integrate multi-source data from RFID, WSN and visual recognition, remove redundancy, extract useful information, improve data quality, transmit the integrated data to central management, form a digital management platform, realize unmanned microbial testing laboratories, enhance laboratory safety and operational efficiency, and reduce energy consumption, in line with green and sustainable development goals.

[0051] In addition, the system uses ultraviolet and ozone sterilization technology, and works in collaboration with the microbial detection robot to automatically sterilize microbial samples, experimental equipment and laboratory environment. It uses the PID control algorithm to dynamically adjust the sterilization parameters to ensure that the sterilization conditions are always in the best state. The PID controller continuously compares the expected environmental parameters with the actual measured values, calculates the control signal, and then adjusts the output of the sterilization equipment, including changing the power of the ultraviolet lamp and controlling the output of the ozone generator to correct deviations and maintain system stability. The algorithm formula is: K p , K i and K d are the proportional coefficient, integral coefficient and differential coefficient respectively, represents the integral of the error over time, is the instantaneous rate of change of the error, that is, the differential of the error, and u(t) is the power and concentration control instructions adjusted at time t to ensure a sterile environment and avoid secondary contamination during the detection process.

[0052] It will be apparent to those skilled in the art that the invention is not limited to the details of the exemplary embodiments described above and that the invention can be implemented in other specific forms without departing from the spirit or essential features of the invention. Therefore, the embodiments should be considered exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description, and it is intended that all variations falling within the meaning and scope of the equivalent elements of the claims be included in the invention. Any reference numeral in a claim should not be considered as limiting the claim to which it relates.

Claims

1. A robotic microbial detection and preparation platform, characterized in that The system consists of the following components: Microbial intelligent detection module: Based on deep learning algorithms, the robot can autonomously learn and adjust strategies according to environmental feedback. It uses behavioral gradient Q learning algorithms and XCSG integrated algorithms to achieve multi-type, small-batch, and multi-batch microbial detection operations, automatically complete sample homogenization, culture medium preparation, microbial incubation, streaking, inoculation, and detection processes, and conduct real-time monitoring and data analysis to achieve unmanned intelligent operation and detection processes. Microbial culture module: provides the environmental conditions required for microbial growth, including temperature, humidity, and light, to ensure the normal growth and reproduction of microorganisms and provide reliable samples for subsequent detection and analysis; Microbial intelligent analysis module: It uses multi-degree-of-freedom robots, servo slides, and flexible grippers, in conjunction with IoT technology, and works with water baths and centrifuges to automatically detect and analyze microbial samples, obtain information on the types and quantities of microorganisms, and achieve full automation of the entire chain from sample processing and detection to result analysis; Data processing module: Analyze and process the test data using big data technology, and use the random forest algorithm model to intelligently analyze the data during the test process, monitor the status of robots and test equipment in real time, and promptly warn of potential quality risks to ensure the accuracy and reliability of the test results; IoT technology integration module: Integrates RFID, WSN and visual recognition technologies to achieve comprehensive digital management of the testing process, including real-time monitoring and data transmission of reagent types, sample data, and equipment operating status, forming an unmanned microbial intelligent testing laboratory to achieve safety and efficiency of testing; Automatic sterilization module: It uses efficient and safe ultraviolet and ozone sterilization technology, and cooperates with the microbial detection robot to automatically sterilize the detected microbial samples, instruments and equipment, and laboratory environment to ensure that no secondary contamination is generated during the detection process and to ensure a sterile environment for the detection process.

2. A robot microorganism detection and preparation platform according to claim 1, characterized in that: The microbial intelligent detection module realizes robot autonomous learning based on a deep learning algorithm. By taking various tentative actions on the perceived environmental state, it obtains the reward and penalty signals of the environmental state for such tentative actions, and continuously adjusts the learning strategy to obtain larger rewards and smaller penalties, which are fed back to the robot to realize rapid convergence of its learning, thereby achieving the purpose of autonomous learning and flexible adaptation to multiple types of detection operations.

3. A robot microorganism detection and preparation platform according to claim 1, characterized in that: The microbial intelligent detection module learns to perform optimal microbial detection operations by interacting with the environment through the behavioral gradient Q learning algorithm. The robot is trained to select the detection strategy for culture medium preparation and microbial incubation conditions based on the environmental feedback of the microbial sample characteristics and the detection equipment status. The behavioral gradient method improves the strategy by directly optimizing the strategy parameters. Q learning is to learn an action value function to indicate the expected return after taking an action in a given state, allowing more direct optimization of the strategy and adapting to the continuous action space. The Q learning update formula is: Among them, Q(s t ,a t ) is in state s at time step t t Take action a t The expected return is α, the learning rate is r t+1 Is to take action a t The reward obtained, γ is the discount factor, which indicates the importance of future rewards relative to immediate rewards, is in the next state s t+1 The robot continuously interacts with the environment and updates the Q table, thereby gradually learning the strategy of choosing the best action under different states.

4. A robot microorganism detection and preparation platform according to claim 3, characterized in that: The microbial intelligent detection module will adopt the behavioral gradient Q learning algorithm and the XCSG integrated algorithm. By combining the robot learning planning rules with the model function under the environmental model, the continuous state and continuous action are integrated with the environment and result feedback to realize sequential decision-making, and the weight matrix value is mapped to XCSG, and the robot learning convergence speed is increased exponentially from the result reward strategy, so as to achieve the purpose of strengthening the robot's autonomous learning to cope with the goal of flexible adaptation of the robot in different experiments and small batch and multi-batch experiments.

5. A robot microorganism detection and preparation platform according to claim 1, characterized in that: The data processing module uses the standardization and mean algorithm in big data technology to analyze and process the microbial detection data. The standardization and mean algorithm formula is: x is the raw microbial data, which integrates data from multiple testing points, multiple batches, and multiple types for efficient storage, query, and analysis. It collects data information on temperature, humidity, microbial types, and quantity during the microbial testing process, monitors the status of robots and testing equipment in real time, promptly identifies potential quality risks, and evaluates the accuracy and reliability of test results.

6. A robot microorganism detection and preparation platform according to claim 5, characterized in that: The data processing module uses a data set processed by big data technology to construct a random forest model, selects feature data from the test data, including key variables such as sample concentration, incubation time, and equipment temperature, as the basis for constructing each tree. For the construction of each decision tree, the algorithm will find the best split point in the selected feature subset to divide the data. This process is repeated so that each tree independently classifies and regresses the samples. The algorithm formula is: T(x) = argmax c∈C ∑ t∈forest I(t(x)=c), T(x) represents the predicted category of the input sample x, C is the set of all categories, t(x) represents the predicted category of the tth tree for the sample x, I is the indicator function, forest represents the entire random forest, and each decision tree is trained through training data. Each decision tree will divide the data according to the selected features to maximize the accuracy of classification and minimize the error of regression. The test data set is used to evaluate the performance of the model, and the trained random forest model is used to predict new data. The predicted results are compared with the actual results, and the accuracy, precision and recall of the model are evaluated. The prediction results are output, and the model is optimized and adjusted according to the evaluation indicators. The prediction results of the random forest model are expressed as the average and voting results of the prediction results of multiple decision trees, that is: F(x) is the final prediction result of the random forest model for sample x, and M is the number of decision trees in the random forest.

7. A robot microorganism detection and preparation platform according to claim 1, characterized in that: The wireless radio frequency identification (RFID) technology in the IoT technology integration module is used to track and manage the reagents and sample items required for microbial detection to ensure their accuracy and traceability. The wireless sensor network (WSN) is used to monitor the environmental conditions of temperature, humidity, and light in the laboratory, as well as the operating status of the detection equipment. By collecting and analyzing these data in real time, it is ensured that the microorganisms are grown and detected in the best environment. Visual recognition technology is used to automatically identify and classify microbial samples.

8. A robot microorganism detection and preparation platform according to claim 7, characterized in that: The IoT technology integration module integrates and processes multi-source data from RFID, WSN and visual recognition through Kalman filtering in the data fusion algorithm, eliminates redundant information, extracts useful microbial data, improves data integrity and accuracy, and recursively combines prior knowledge and observation data to update the estimation of the system state. The algorithm formula of the Kalman filter prediction and update steps is: Prediction: renew: in, represents state estimation, F is the state transfer matrix, B is the control input influence matrix, u is the control input, K is the Kalman gain, z is the measurement value, and H is the observation matrix. Different multi-source data are integrated, including the microbial detection identity and location information provided by RFID, the environmental parameters transmitted by WSN, and the image and video information captured by the visual recognition system. Intelligent fusion improves data quality and provides a more comprehensive and accurate information basis for decision-making.

9. A robot microorganism detection and preparation platform according to claim 1, characterized in that: After the experiment is completed, the ultraviolet sterilization technology in the automatic sterilization module destroys the DNA structure of the microorganisms, and the ultraviolet lamp is used to generate ultraviolet radiation to irradiate the sterilized microbial samples, instruments and equipment and the environment, thereby preventing their replication and reproduction, thereby achieving the sterility effect of the equipment. The ozone sterilization technology reacts with the cell wall, cell membrane and enzymes in the cells of the microorganisms to destroy their structure, and ozone is generated by the ozone generator and transported to the area that needs to be sterilized, thereby achieving the sterility effect of the equipment.

10. A robot microorganism detection and preparation platform according to claim 9, characterized in that: The automatic sterilization module automatically adjusts the sterilization parameters through the PID control algorithm to ensure that the temperature, humidity, ultraviolet intensity and ozone concentration are maintained within the optimal range to achieve the best sterilization effect. The PID controller continuously compares the expected environmental parameters with the actual measured values, calculates the control signal, and then adjusts the output of the sterilization equipment, including changing the power of the ultraviolet lamp and controlling the output of the ozone generator to correct the deviation and maintain the stability of the system. The algorithm formula is: K p , K i and K d are the proportional coefficient, integral coefficient and differential coefficient respectively, represents the integral of the error over time, is the instantaneous rate of change of the error, that is, the differential of the error, and u(t) is the power and concentration control instructions adjusted at time t.

Images