Real-time Monitoring System and Method for Recyclable Packaging Appliances Based on Multi-Sensor Fusion

Abstract

The present invention discloses a real-time monitoring system and method for recyclable packaging appliances based on multi-sensor fusion, which relates to the field of Internet of Things technology. By collecting historical sensing data, the observed values #imgabs1# of the acceleration sensor and the observed values #imgabs2# of the vibration sensor at the moment #imgabs0# are obtained, the motion states of the recyclable packaging appliances to be monitored are defined, the probability values when the recyclable packaging appliances to be monitored are in the #imgabs3#-th motion state are calculated, the first probability distribution of the observed values #imgabs4# of the acceleration sensor and the second probability distribution of the observed values #imgabs5# of the vibration sensor in different motion states are constructed, according to the obtained probability values, the first probability distribution and the second probability distribution, the third probability distribution corresponding to the observed values #imgabs7# of the acceleration sensor and the observed values #imgabs8# of the vibration sensor collected at the moment #imgabs6# when in the #imgabs9#-th motion state is calculated, and according to the calculation results of the obtained third probability distribution, the motion state of the recyclable packaging appliances to be monitored at the moment #imgabs10# is determined.

Description

Technical Field

[0001] The present invention relates to the technical field of the Internet of Things, and particularly to a real-time monitoring system and method for recyclable packaging appliances based on multi-sensor fusion. Background Art

[0002] With the rapid development of the logistics and supply chain industries, the application of recyclable packaging appliances in transportation and storage has become increasingly widespread. In modern logistics and supply chain management, the real-time monitoring of recyclable packaging appliances is of great significance for improving transportation efficiency, ensuring the safety of goods, and optimizing logistics costs.

[0003] However, there are many deficiencies in the management and monitoring of recyclable packaging appliances in the existing technologies. The main manifestations are as follows: Traditional methods mainly rely on manual inspections or single-sensor technologies, and it is difficult to achieve all-round real-time monitoring of packaging appliances. There are the following technical problems: 1) Insufficient monitoring accuracy: A single sensor can only provide limited information and it is difficult to comprehensively reflect the movement state of the packaging appliance; 2) Weak anti-interference ability: In a complex transportation and storage environment, a single sensor is easily affected by external interference, which affects the monitoring results and leads to misjudgments; 3) Unable to provide real-time feedback: There are delays in data fusion and processing in the existing technologies, and it is impossible to achieve real-time monitoring of the state of packaging appliances.

[0004] At present, multi-sensor fusion technology has been applied in fields such as autonomous driving and industrial automation in the existing technologies, but it has not been fully developed and utilized in the monitoring system of recyclable packaging appliances. Therefore, how to achieve all-round real-time monitoring of recyclable packaging appliances through multi-sensor fusion technology and accurately determine their states is a technical problem that needs to be solved urgently at present. For this reason, we propose a real-time monitoring system and method for recyclable packaging appliances based on multi-sensor fusion. Summary of the Invention

[0005] The main purpose of the present invention is to provide a real-time monitoring system and method for recyclable packaging appliances based on multi-sensor fusion, which can effectively solve the problems in the background art.

[0006] To achieve the above purpose, the technical solution adopted by the present invention is as follows:

[0007] A real-time monitoring method for recyclable packaging appliances based on multi-sensor fusion, comprising:

[0008] Step 1: Collect historical sensing data of the recyclable packaging appliance to be monitored within the circulation period T at a fixed sampling frequency λ, and obtain the observation value of the acceleration sensor at the moment and the observation value of the vibration sensor , where ∈T;

[0009] Step 2: Define the motion states of the recyclable packaging appliances to be monitored as ={ , , , }, and calculate the probability value when the recyclable packaging appliances to be monitored are in the th motion state based on the acquired historical sensing data;

[0010] Step 3: Construct the first probability distribution of the observed values of the acceleration sensor and the second probability distribution of the observed values of the vibration sensor for different motion states respectively according to the acquired historical sensing data;

[0011] Step 4: Calculate the third probability distribution corresponding to the motion state when the observed values of the acceleration sensor and the observed values of the vibration sensor collected at the moment are based on the acquired probability value , the first probability distribution and the second probability distribution , and the calculation formula is: = × × × ;

[0012] Step 5: Determine the motion state of the recyclable packaging appliances to be monitored at the moment according to the calculation result of the acquired third probability distribution

[0013] The real-time monitoring system for recyclable packaging appliances based on multi-sensor fusion includes:

[0014] The historical sensing data acquisition module is used to acquire the historical sensing data of the recyclable packaging appliances to be monitored within the circulation period T at a fixed sampling frequency λ, and obtain the observed values of the acceleration sensor and the observed values of the vibration sensor at the moment;

[0015] The motion state definition module is used to define the motion state of the recyclable packaging appliances to be monitored, where the motion state is ={ , , , };​

[0016] A data processing module, configured to calculate the probability value when the recyclable packaging appliance to be monitored is in the th motion state based on the acquired historical sensing data ;

[0017] A first probability distribution calculation module, configured to respectively construct the first probability distribution of the observed values of the acceleration sensor in different motion states based on the acquired historical sensing data ; ;

[0018] A second probability distribution calculation module, configured to respectively construct the second probability distribution of the observed values of the vibration sensor in different motion states based on the acquired historical sensing data ; ;

[0019] A real-time data acquisition module, configured to acquire the observed values of the acceleration sensor and the observed values of the vibration sensor collected by the recyclable packaging appliance to be monitored at the moment and ;

[0020] A third probability distribution calculation module, configured to calculate the third probability distribution corresponding to the observed values of the acceleration sensor and the observed values of the vibration sensor collected at the moment when in the th motion state based on the acquired probability value , the first probability distribution and the second probability distribution , and the calculation formula is: = × × × × ;

[0021] A motion state evaluation module, configured to determine the motion state of the recyclable packaging appliance to be monitored at the moment according to the calculation result of the acquired third probability distribution ;

[0022] When is the maximum value among all the calculation results, it is determined that the recyclable packaging appliance to be monitored is most likely to be in the th motion state at the moment;

[0023] When is the minimum value among all the calculation results, it is determined that at At a certain moment, the probability that the reusable packaging appliance to be monitored is in the th motion state is the smallest.

[0024] The system further includes a memory, a processor, and a computer program stored on the memory and executable on the processor.

[0025] Further, in step two, the classification principle of the motion state of the reusable packaging appliance to be monitored is:

[0026] When = it indicates that the motion state of the reusable packaging appliance to be monitored is a stationary state;

[0027] When = it indicates that the motion state of the reusable packaging appliance to be monitored is a uniform motion state;

[0028] When = it indicates that the motion state of the reusable packaging appliance to be monitored is a variable-speed motion state;

[0029] When = it indicates that the motion state of the reusable packaging appliance to be monitored is a state of being impacted.

[0030] Further, in step two, the probability value that the reusable packaging appliance to be monitored is in the th motion state is calculated by the formula:

[0031] =

[0032] In the formula, represents the sampling frequency of the reusable packaging appliance to be monitored in the th motion state in the historical sensing data; = 1, 2, 3, 4.

[0033] Further, in step three, the first probability distribution of the observed value of the acceleration sensor is calculated by the formula:

[0034] = =

[0035] In the formula, is the probability density function subject to the Gaussian distribution; represents being in the Observation values of the acceleration sensor in a certain motion state The mean value; Denoted as the observation value of the acceleration sensor in the certain motion state The variance; π is the pi.

[0036] Furthermore, in step three, the observation value of the vibration sensor The second probability distribution The calculation formula is:

[0037] = =

[0038] In the formula, Is the probability density function that follows the Gaussian distribution; Denoted as the observation value of the vibration sensor in the certain motion state The mean value; Denoted as the observation value of the vibration sensor in the certain motion state The variance; π is the pi.

[0039] Furthermore, in step five, at The determination principle of the motion state of the recyclable packaging appliance to be monitored at the moment is:

[0040] For = 1, 2, 3, 4;

[0041] When Is the maximum value among all the calculation results, it is determined that the recyclable packaging appliance to be monitored at The moment is most likely to be in the certain motion state;

[0042] When Is the minimum value among all the calculation results, it is determined that the recyclable packaging appliance to be monitored at The moment is least likely to be in the certain motion state.

[0043] The present invention has the following beneficial effects

[0044] Compared with the prior art, the technical solution of the present invention acquires the historical sensing data of the recyclable packaging appliance to be monitored within the circulation period T at a fixed sampling frequency λ, and obtains The observation value of the acceleration sensor at the moment And the observation value of the vibration sensor , define the motion state of the recyclable packaging appliance to be monitored. According to the acquired historical sensing data, calculate the probability value when the recyclable packaging appliance to be monitored is in the th motion state, construct the first probability distribution of the observed values of the acceleration sensor and the second probability distribution of the observed values of the vibration sensor under different motion states. According to the acquired probability value, the first probability distribution and the second probability distribution, calculate the third probability distribution corresponding to the observed values of the acceleration sensor and the observed values of the vibration sensor collected at the moment when the recyclable packaging appliance to be monitored is in the th motion state, and determine the motion state of the recyclable packaging appliance to be monitored at the moment according to the calculation result of the acquired third probability distribution, realizing the all-round real-time monitoring of the recyclable packaging appliance and accurately determining its state, effectively solving the technical problems such as insufficient monitoring accuracy of a single sensor and weak anti-interference ability in complex transportation and storage environments in the prior art. Brief Description of the Drawings

[0045] Figure 1 is a schematic flowchart of the real-time monitoring method for recyclable packaging appliances based on multi-sensor fusion of the present invention;

[0046] Figure 2 is a schematic structural diagram of the real-time monitoring system for recyclable packaging appliances based on multi-sensor fusion of the present invention. Detailed Embodiments

[0047] The following further describes the present invention in conjunction with specific embodiments. Among them, the drawings are only for illustrative purposes, showing only schematic diagrams, not physical diagrams, and should not be construed as a limitation to the present invention. In order to better illustrate the specific embodiments of the present invention, some components in the drawings will be omitted, enlarged or reduced, and do not represent the size of the actual product.

[0048] The implementation process of the technical solution of the present invention includes the following steps:

[0049] Step 1: Collect the historical sensing data of the recyclable packaging appliance to be monitored within the circulation period T at a fixed sampling frequency λ, and obtain the observed values of the acceleration sensor and the observed values of the vibration sensor at the moment, where ∈T;

[0050] It should be noted that for different sensor data, this solution mainly involves acceleration sensors and vibration sensors. When collecting sensor data, time synchronization processing is required. Specifically, a method combining hardware synchronization and software synchronization can be adopted.

[0051] 1) Hardware synchronization

[0052] Hardware synchronization provides the same reference time for each sensor through a unified clock source, thus achieving time synchronization at the hardware level. Common hardware synchronization methods include:

[0053] GPS time synchronization: Utilize the high-precision time signal provided by GPS (such as PPS+NMEA) as the unified time source, and each sensor calibrates its own clock according to the GPS time.

[0054] PTP protocol (IEEE 1588): Achieve sub-microsecond clock synchronization between multiple sensors and the host through Ethernet. The PTP protocol calculates the network transmission delay and clock deviation through the interaction of synchronization messages between the master and slave devices, thus achieving high-precision time synchronization.

[0055] 2) Software synchronization

[0056] Software synchronization aligns sensor data in time through software algorithms, mainly using timestamps for matching. Common software synchronization methods include:

[0057] Timestamp-based matching: Unify the data of each sensor to the sensor data with a longer scan period (lower frequency). For example, if the sampling frequency of the lidar is 12.5Hz and the sampling frequency of the camera is 30Hz, then based on the sampling frequency of the lidar, find the camera data closest to the lidar sampling moment.

[0058] Interpolation method: For sensors with inconsistent sampling frequencies, calculate the equivalent data at a certain moment through an interpolation algorithm. For example, according to the radar data before and after the camera sampling moment, calculate the radar data corresponding to the camera sampling moment through linear interpolation.

[0059] 3) Combination of hardware synchronization and software synchronization

[0060] In practical applications, hardware synchronization and software synchronization are usually combined to achieve more efficient time synchronization. For example:

[0061] Hardware synchronization provides the reference time: Provide a unified time reference for the sensors through the GPS or PTP protocol.

[0062] Software synchronization processes time deviation: On the basis of hardware synchronization, further align the data of each sensor through software algorithms to ensure the temporal consistency of the data.

[0063] Through the above steps, different sensor data can be synchronized in space and time to reflect the true state of the reusable packaging appliance to be monitored at time t.

[0064] Step 2: Define the motion state of the reusable packaging appliance to be monitored as ={ , , , }, where when = , it indicates that the motion state of the reusable packaging appliance to be monitored is a stationary state; when = , it indicates that the motion state of the reusable packaging appliance to be monitored is a uniform motion state; when = , it indicates that the motion state of the reusable packaging appliance to be monitored is a variable motion state; when = , it indicates that the motion state of the reusable packaging appliance to be monitored is a state of being impacted.

[0065] Through the above steps, the motion states of the reusable packaging appliance to be monitored can be classified, facilitating subsequent assessment of the possibility of the motion state of the reusable packaging appliance to be monitored. The specific definition method can adopt the artificial calibration method, where historical sensing data is calibrated manually to form datasets under different motion states.

[0066] Step 3: According to the obtained historical sensing data, calculate the probability value when the reusable packaging appliance to be monitored is in the th motion state. The calculation formula is:

[0067] =

[0068] In the formula, represents the sampling frequency of the reusable packaging appliance to be monitored in the th motion state in the historical sensing data; = 1, 2, 3, 4.

[0069] Step 4: Construct the first probability distribution of the observed values of the acceleration sensor under different motion states according to the obtained historical sensing data. The calculation formula is:

[0070] = =

[0071] In the formula, is the probability density function that follows the Gaussian distribution; represents the mean value of the observed values of the acceleration sensor in the th motion state; of; represents the variance of the observed values of the acceleration sensor in the th motion state; of; is the pi;

[0072] It should be noted that in this solution, it is assumed that the observed values of the acceleration sensor follow the Gaussian distribution. Specifically, it is necessary to test the observed values of the collected acceleration sensor for confirmation. The test method is as follows: Through hypothesis testing (such as the Shapiro-Wilk test) and frequency distribution histogram analysis, it can be judged whether the acceleration data follows the Gaussian distribution.

[0073] Step 5: Construct the second probability distribution of the observed values of the vibration sensor in different motion states according to the obtained historical sensing data , and the calculation formula is: , the calculation formula is:

[0074] = =

[0075] In the formula, is the probability density function that follows the Gaussian distribution; represents the mean value of the observed values of the vibration sensor in the th motion state; of; represents the variance of the observed values of the vibration sensor in the th motion state; of; is the pi;

[0076] It should be noted that the observed values of the vibration sensor also need to be tested to confirm whether they follow the Gaussian distribution. The test method is the same as that in Step 4 above and will not be elaborated here.

[0077] Step 6: Collect the observed values of the acceleration sensor of the recyclable packaging appliance to be monitored at the moment and the observed values of the vibration sensor.

[0078] Step 7: According to the obtained probability value , the first probability distribution and the second probability distribution , calculate when at Observation values of the acceleration sensor collected at a moment and the observation values of the vibration sensor corresponding to the third probability distribution in the th motion state , and the calculation formula is: = × × ;

[0079] Step 8: According to the obtained third probability distribution calculation results, determine the motion state of the recyclable packaging appliance to be monitored at moment; The determination principle of the motion state is: for = 1, 2, 3, 4;

[0080] When is the maximum value among all calculation results, it is determined that the recyclable packaging appliance to be monitored at moment is most likely to be in the th motion state;

[0081] When is the minimum value among all calculation results, it is determined that the recyclable packaging appliance to be monitored at moment is least likely to be in the th motion state.

[0082] Specifically, for = 1, 2, 3, 4; The calculation results of the third probability distribution include , , , ;

[0083] If there is > > > , then at moment, the recyclable packaging appliance to be monitored is most likely to be in the th motion state and least likely to be in the th motion state, that is, most likely to be in a static state and least likely to be in a state of being impacted;

[0084] Similarly, if there is < < < , then at moment, the recyclable packaging appliance to be monitored is least likely to be in the th motion state and most likely to be in the The probability of a certain motion state is the highest, that is, the probability of being in a stationary state is the lowest, and the probability of being in a state of being impacted is the highest.

[0085] Step 9: Dynamic update

[0086] As the observed values of the acceleration sensor and the vibration sensor are continuously collected and updated, the third probability distribution can be recursively updated, thereby realizing the real-time estimation of the device motion state.

[0087] Through the above technical solution, it is possible to achieve all-round real-time monitoring of the recyclable packaging appliance and accurately determine its state, thereby effectively solving the technical problems such as insufficient monitoring accuracy of a single sensor and weak anti-interference ability in complex transportation and storage environments in the prior art.

[0088] The above shows and describes the basic principles, main features and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited by the above embodiments. What is described in the above embodiments and the specification only illustrates the principles of the present invention. Without departing from the spirit and scope of the present invention, the present invention will have various changes and improvements, and these changes and improvements all fall within the scope of the present invention claimed. The scope of protection claimed by the present invention is defined by the appended claims and their equivalents.

Claims

1. A real-time monitoring method for recyclable packaging appliances based on multi-sensor fusion, characterized in that, Including: Step 1: Collect historical sensing data of the reusable packaging appliance to be monitored within the circulation period T at a fixed sampling frequency λ to obtain the observed value of the acceleration sensor at the moment and the observed value of the vibration sensor , where ∈T; Step 2: Define the motion state of the recyclable packaging appliance to be monitored as ={ , , , }, and calculate the probability value when the recyclable packaging appliance to be monitored is in the th motion state based on the obtained historical sensing data; Step 3: Construct the first probability distribution of the observed values of the acceleration sensor and the second probability distribution of the observed values of the vibration sensor under different motion states respectively according to the obtained historical sensing data of the acceleration sensor and the second probability distribution of the observed values of the vibration sensor ; Step 4: According to the obtained probability values , the first probability distribution and the second probability distribution , calculate the third probability distribution corresponding to the observed value of the acceleration sensor and the observed value of the vibration sensor collected at the moment in the th motion state. The calculation formula is: = × × ; Step 5: Determine the motion state of the recyclable packaging appliance to be monitored at the moment according to the calculation result of the obtained third probability distribution.

2. The real-time monitoring method of the recyclable packaging appliance based on multi-sensor fusion according to claim 1, characterized in that, In step two, the classification principle of the motion state of the reusable packaging appliance to be monitored is as follows: When = it indicates that the motion state of the reusable packaging appliance to be monitored is the static state; When = it indicates that the motion state of the reusable packaging appliance to be monitored is a uniform motion state; When = it indicates that the motion state of the reusable packaging appliance to be monitored is a variable-speed motion state; When = it indicates that the motion state of the reusable packaging appliance to be monitored is in a state of being impacted.

3. The real-time monitoring method for recyclable packaging appliances based on multi-sensor fusion according to claim 1, wherein In step two, the probability value of the reusable packaging appliance to be monitored being in the th motion state is calculated by the formula: = In the formula, represents the sampling frequency of the recyclable packaging appliance to be monitored in the th motion state in the historical sensing perception data; = 1, 2, 3, 4.

4. The real-time monitoring method of the recyclable packaging appliance based on multi-sensor fusion according to claim 1, characterized in that, In step three, the observed value of the acceleration sensor The first probability distribution of The calculation formula is: = = Wherein, is the probability density function that follows the Gaussian distribution; represents the mean value of the observed value of the acceleration sensor in the th motion state; ; represents the variance of the observed value of the acceleration sensor in the th motion state; ; is the pi.

5. The real-time monitoring method of the recyclable packaging appliance based on multi-sensor fusion according to claim 1, characterized in that In step three, the observed value of the vibration sensor The second probability distribution of The calculation formula is: = = Wherein, is the probability density function obeying the Gaussian distribution; represents the mean value of the observed value of the vibration sensor in the th motion state; ; represents the variance of the observed value of the vibration sensor in the th motion state; ; is the pi.

6. The real-time monitoring method of the recyclable packaging appliance based on multi-sensor fusion according to claim 1, characterized in that, In step five, at the determination principle for the motion state of the reusable packaging appliance to be monitored at that moment is as follows: For = 1, 2, 3, 4; When is the maximum value among all calculation results, it is determined that the recyclable packaging appliance to be monitored is most likely in the th motion state at the moment; When is the minimum value among all calculation results, it is determined that the recyclable packaging appliance to be monitored is least likely to be in the th motion state at the moment.

7. Real-time monitoring system for recyclable packaging appliances based on multi-sensor fusion, characterized in that Including: A historical sensing data acquisition module, which is used to collect historical sensing data of the reusable packaging appliance to be monitored within a circulation period T at a fixed sampling frequency λ, and obtain the observed value of the acceleration sensor at the moment and the observed value of the vibration sensor , where ∈T; A motion state definition module, which is used to define the motion state of a recyclable packaging appliance to be monitored. Among them, the motion state is ={ , , , }, when = , it indicates that the motion state of the recyclable packaging appliance to be monitored is a static state; when = , it indicates that the motion state of the recyclable packaging appliance to be monitored is a uniform motion state; when = , it indicates that the motion state of the recyclable packaging appliance to be monitored is a variable-speed motion state; when = , it indicates that the motion state of the recyclable packaging appliance to be monitored is a state of being impacted; A data processing module, configured to calculate a probability value when the recyclable packaging appliance to be monitored is in the th motion state according to the acquired historical sensing data , and the calculation formula is: = ; where, represents the sampling frequency when the recyclable packaging appliance to be monitored in the historical sensing data is in the th motion state; = 1, 2, 3, 4; The first probability distribution calculation module is used to construct the first probability distribution of the observed values of the acceleration sensor in different motion states according to the acquired historical sensing data of , and the calculation formula is: = = , where is the probability density function subject to the Gaussian distribution; represents the mean value of the observed value of the acceleration sensor in the th motion state; represents the variance of the observed value of the acceleration sensor in the th motion state; is pi; The second probability distribution calculation module is used to construct the second probability distribution of the observed values of the vibration sensor under different motion states according to the obtained historical sensing data of . The calculation formula is as follows: = = ; In the formula, is the probability density function subject to the Gaussian distribution; represents the mean value of the observed value of the vibration sensor in the th motion state; represents the variance of the observed value of the vibration sensor in the th motion state; is the pi; A real-time data acquisition module for collecting the observed values of an acceleration sensor and a vibration sensor of a reusable packaging appliance to be monitored at the moment of collection and the observed values of the vibration sensor ; A third probability distribution calculation module, which is used to calculate, according to the obtained probability values , the first probability distribution and the second probability distribution , the third probability distribution corresponding to the observation value of the acceleration sensor and the observation value of the vibration sensor collected at the moment in the th motion state, and the calculation formula is: = × × = × × ; A motion state evaluation module, which is used to determine the motion state of a recyclable packaging appliance to be monitored at a certain moment according to the calculation result of the obtained third probability distribution, where the determination principle of the motion state is: for i = 1, 2, 3, 4; ​ When is the maximum value among all calculation results, it is determined that the recyclable packaging appliance to be monitored is most likely in the th kind of motion state at the When is the minimum value among all calculation results, it is determined that the recyclable packaging appliance to be monitored is least likely to be in the th kind of motion state at the 8. The real-time monitoring system for recyclable packaging appliances based on multi-sensor fusion according to claim 7, characterized in that, The system further includes a memory, a processor, and a computer program stored on the memory and executable on the processor. Among them, when the processor executes the program, it can implement the steps of the real-time monitoring method of the reusable packaging appliance based on multi-sensor fusion described in any one of claims 1-6.

Images