Intelligent logistics monitoring method and system based on Internet of Things

Abstract

The invention discloses an intelligent logistics monitoring method and system based on the Internet of Things, and relates to the technical field of intelligent logistics monitoring, and the method comprises the steps that a sensor carries out the self-inspection and collects data; the local controller collects sensor data and transmits the sensor data to the edge gateway for primary checking; local detection is carried out after primary checking is completed, and data are fused and uploaded to a cloud platform; and the cloud platform performs secondary check, performs anomaly detection on the data, generates an early warning, and feeds back the data to the edge gateway for scheduling. According to the method, the security and reliability of the data are ensured, and data loss caused by a single-point fault is avoided. Dynamic anomaly detection is realized, and the intelligent level of logistics monitoring is improved. The distribution scheme can be optimized in real time, and the efficiency and stability of logistics transportation are improved. AES encryption and HMAC verification modes are adopted, and data are effectively prevented from being tampered in the transmission process.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent logistics monitoring, and specifically to an intelligent logistics monitoring method and system based on the Internet of Things. Background Art

[0002] In recent years, the Internet of Things technology has developed rapidly and has been widely applied in the field of intelligent logistics. With the continuous progress of key technologies such as sensors, edge computing, and cloud platforms, traditional logistics monitoring systems have gradually evolved towards intelligence and real-time. The application of various temperature and humidity, vibration, positioning, and RFID sensors in transport vehicles, warehousing centers, and distribution links has become increasingly mature, providing rich means for data collection. At the same time, the mode of local controllers and edge gateways collaborating to process data has gradually emerged, realizing preliminary data fusion and local anomaly detection, effectively alleviating the delay and security problems in data transmission. The introduction of the cloud platform has further promoted the management and analysis of full-link data, enabling higher levels of intelligence in links such as anomaly detection, early warning generation, and dispatching instruction issuance. Currently, hierarchical data processing and multi-level encrypted transmission technologies are commonly used in intelligent logistics monitoring systems at home and abroad, but the overall system still has deficiencies in terms of real-time performance, data integrity, and anomaly detection accuracy. Especially in the face of a dynamically changing logistics environment, traditional solutions are difficult to achieve fast and accurate multi-level verification and intelligent scheduling of data.

[0003] In the prior art, after sensor data is collected, it is directly uploaded to the cloud platform, often ignoring the importance of multiple checks and adaptive detection at the local level, resulting in the data being easily interfered with or tampered with during transmission, and the response speed of anomaly detection algorithms being relatively low, making it difficult to meet the real-time requirements of full-link logistics monitoring. Existing systems rely mainly on a single encryption method for data security and do not fully combine multiple technical means such as AES, HMAC, and digital signatures, thus lacking a perfect security guarantee in all aspects of data transmission. On the other hand, edge gateways only perform simple data caching and preliminary fusion, and do not effectively utilize historical data for intelligent threshold adjustment, making it impossible to respond in a timely manner to sudden anomalies such as temperature and humidity, and vibration in the logistics environment. Summary of the Invention

[0004] In view of the above problems, the present invention is proposed.

[0005] Therefore, the technical problems solved by the present invention are: problems such as unstable transmission, insufficient real-time performance, and slow anomaly response existing in the existing logistics monitoring system.

[0006] To solve the above technical problems, the present invention provides the following technical solutions: A smart logistics monitoring method based on the Internet of Things, including: sensors perform self-checks and collect data; the local controller collects sensor data and transmits it to the edge gateway for a primary verification; after the primary verification is completed, local detection is carried out, the data is fused and uploaded to the cloud platform; the cloud platform performs a secondary verification, detects anomalies in the data and generates an early warning, and feeds back to the edge gateway for scheduling.

[0007] As a preferred solution of the smart logistics monitoring method based on the Internet of Things according to the present invention, wherein: the self-check of the sensors includes that the sensors include temperature and humidity sensors, vibration sensors, positioning modules and RFID; before the transport vehicle departs, the sensors in the vehicle perform self-checks to confirm that the temperature, humidity, GPS and RFID devices are operating normally; after the vehicle starts, a sampling period is set using timer interrupts, and temperature, humidity, vibration, GPS and RFID data are collected at preset intervals; the sensors convert analog data into digital signals, attach a timestamp and a device number, and then transmit the data to the local controller.

[0008] As a preferred solution of the smart logistics monitoring method based on the Internet of Things according to the present invention, wherein: the collection of sensor data by the local controller includes, when the device is started, obtaining a shared key through a secure channel; after each data collection, first call the AES encryption function to encrypt the data packet, then calculate the HMAC value, and append the HMAC to the end of the data packet; the receiving end verifies the HMAC before decryption to ensure that the data has not been tampered with during transmission; the local controller performs preliminary noise reduction and correction on the data; after the local controller collects the data of multiple sensors in the vehicle, it packages them into a standard data packet; the local controller encrypts the data packet and adds a digital signature, and prepares to upload; after the edge gateway establishes a connection with each local controller through short-distance communication, performs identity authentication and exchanges keys through a handshake, the data packet is uploaded to the edge gateway.

[0009] As a preferred solution of the intelligent logistics monitoring method based on the Internet of Things according to the present invention, wherein: the edge gateway performs a primary verification, which includes decrypting and performing HMAC verification on the received data packet first, and then aligning the data using the timestamp; the edge gateway decrypts and verifies the data, performs a primary verification, and judges the device identity and data integrity; fuses the data within the same time period, the same vehicle, and the same warehouse, supplements the GPS positioning data, associates the temperature, humidity, and vibration data with the current location, and stores a cache backup locally; embeds an adaptive judgment module in the edge gateway for local detection, and sets the environmental anomaly threshold according to the preset rules in the initial state; the adaptive judgment module records the historical data and local anomaly events, and uses a lightweight reinforcement learning algorithm and a parameter adaptive update mechanism of the synchronous cloud platform to dynamically adjust the local anomaly detection threshold and judgment criteria according to the latest policy parameters fed back by the cloud. The local detection includes that when the edge gateway identifies abnormal temperature and humidity, it immediately displays a warning on the local controller of the in-vehicle terminal and sends an adjustment instruction to the local controller to adjust the temperature and humidity; if it cannot be adjusted, it transmits the abnormal data to the cloud platform, and the cloud platform predicts the temperature control change trend and issues an adjustment instruction, which is fed back to the vehicle terminal via the edge gateway to complete the parameter adjustment; if it still cannot be adjusted, the cloud platform starts to adjust the delivery plan. The adjustment of the delivery plan includes adjusting the travel route when there are risks of delay and goods; the GPS, RFID, and sensors in the delivery vehicle collect data in real time, and the edge gateway integrates the vehicle driving trajectory, road condition information, and the collected data and sends them to the cloud platform through a wireless network; the cloud platform performs real-time analysis on the uploaded data and optimizes the delivery route in combination with traffic information; when there is a risk of delivery delay, the cloud platform issues a warning and feeds back the new delivery plan to the driver and the dispatching center; when there is a goods risk, the cloud platform issues a warning and plans the nearest distribution center and warehousing center for emergency dispatching.

[0010] As a preferred solution of the intelligent logistics monitoring method based on the Internet of Things according to the present invention, wherein: the process of fusing the data and uploading it to the cloud platform includes that the edge gateway prepares to upload the fused data to the cloud platform; packs the cache data into messages conforming to the MQTT protocol format, with QoS settings attached; embeds device identification, timestamp, data type, data content, and checksum information in each message, and then performs encryption processing; the data packet is transmitted to the cloud platform through relay devices during the transmission process, and each relay device performs data verification and then forwards it.

[0011] As a preferred solution of the Internet of Things-based intelligent logistics monitoring method described in the present invention, wherein: the secondary verification by the cloud platform includes that the cloud receiving module continuously listens for data uploads from the edge gateway; after the cloud platform receives the data, it performs secondary verification, including digital signature verification, integrity check, timestamp comparison, and detects whether there is abnormal data; when abnormal data is detected, a warning record is generated and sent to the edge gateway for specific scheduling; and it is pushed to the scheduling center and in-vehicle terminal through the message queue; after the secondary verification is completed, the uploaded data is stored in the distributed database and data lake according to the preset classification rules, and at the same time, corresponding log records are generated.

[0012] As a preferred solution of the Internet of Things-based intelligent logistics monitoring method described in the present invention, wherein: the abnormal detection includes that after the data is stored in the database, the background data processing module is immediately started; after the cloud RL module obtains the full-link data, it calculates rewards and punishments with the actual scheduling effect and abnormal handling situation as the feedback content; the feedback content includes the actual time taken for the transportation path, the abnormal warning accuracy rate, and the operation indicators after the execution of the scheduling instruction; based on the feedback results, the cloud RL model continuously updates the strategy online, and outputs the optimized scheduling, path, and abnormal detection criteria in real time, and sends the update results to each edge gateway through a secure channel; after receiving the update parameters, the edge gateway immediately adjusts the parameters of the local adaptive detection model.

[0013] As a preferred solution of the Internet of Things-based intelligent logistics monitoring method described in the present invention, wherein: the abnormal detection includes that after the data is stored in the cloud platform, the background data processing module is immediately started; after the cloud RL module obtains the full-link data, it calculates rewards and punishments with the actual scheduling effect and abnormal handling situation as the feedback content; the feedback content includes the actual time taken for the transportation path, the abnormal warning accuracy rate, and the operation indicators after the execution of the scheduling instruction; based on the feedback results, the cloud RL model continuously updates the strategy online, and outputs the optimized scheduling, path, and abnormal detection criteria in real time, and sends the update results to each edge gateway through a secure channel; after receiving the update parameters, the edge gateway immediately adjusts the parameters of the local adaptive detection model.

[0014] As a preferred solution of the Internet of Things-based intelligent logistics monitoring system described in the present invention, wherein: it includes a sensor module, a local control module, an edge gateway module, and a cloud platform module; The sensor module, after self-checking, collects temperature, humidity, vibration, GPS, and RFID data and sends it to the local controller; The local control module includes a local controller, which encrypts the data with AES, calculates HMAC, and performs digital signature. After completing the security encapsulation, it sends the device online confirmation and data packet to the edge gateway through a secure channel; After receiving the data, the edge gateway module decrypts, verifies, and performs a primary check; aligns and fuses the data using timestamps, and the local adaptive detection module detects anomalies and gives on-site feedback warnings; packs the data into MQTT messages and uploads them to the cloud platform via a wireless network; The cloud platform module is used to perform a secondary check after receiving the data packets, generate warnings and issue scheduling instructions. At the same time, the cloud RL module provides feedback optimization for the full-link data.

[0015] A computer device includes a memory and a processor. The memory stores a computer program, and when the processor executes the computer program, the steps of the intelligent logistics monitoring method based on the Internet of Things are implemented.

[0016] A computer-readable storage medium stores a computer program thereon. When the computer program is executed by a processor, the steps of the intelligent logistics monitoring method based on the Internet of Things are implemented.

[0017] Advantages of the present invention: The method of the present invention ensures the security and reliability of data, and avoids data loss caused by single-point failures. It realizes dynamic anomaly detection and improves the intelligent level of logistics monitoring. It can instantly optimize the distribution plan, improve the efficiency and stability of logistics transportation. By using the AES encryption and HMAC verification methods, it effectively prevents data from being tampered with during transmission. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0019] Figure 1 It is the overall flowchart of an intelligent logistics monitoring method based on the Internet of Things provided by the first embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0020] To make the above objects, features, and advantages of the present invention more obvious and understandable, the following will provide a detailed description of the specific embodiments of the present invention with reference to the accompanying drawings of the specification. Obviously, the described embodiments are 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 shall fall within the protection scope of the present invention.

[0021] Example 1, refer to Figure 1, which is an embodiment of the present invention, provides an Internet of Things-based intelligent logistics monitoring method, including: S1: The sensor performs self-check and collects data.

[0022] Further, the self-check of the sensor includes that the sensor includes a temperature and humidity sensor, a vibration sensor, a positioning module, and an RFID; before the transport vehicle departs, the sensors in the vehicle perform self-check to confirm that the temperature, humidity, GPS, and RFID devices are operating normally.

[0023] The local controller establishes a connection with each sensor and reads the device status information; the local controller sends a device online status confirmation message to the edge gateway for identity authentication and handshake to exchange keys.

[0024] After the vehicle starts, a sampling period is set using a timer interrupt, and temperature, humidity, vibration, GPS, and RFID data are collected every preset period; the sensor converts the analog data into a digital signal and attaches a timestamp and a device number before transmitting it to the local controller.

[0025] Before the transport vehicle departs, the local controller sequentially sends a "self-check command" to each sensor through interfaces such as I²C, SPI, and UART. After each sensor receives the command, it starts the self-check program and returns the self-check result to the local controller. If the status codes of all sensors are within the "normal" range, the local controller considers that the temperature, humidity, GPS, and RFID devices are all operating normally; otherwise, the system displays an error prompt on the LCD display or the in-vehicle terminal.

[0026] It should be noted that after collection, the local controller uses the AES-128 / 256 encryption algorithm to encrypt the data packet. Before encryption, the controller calls the internal library function to generate a data packet digest and calculates the HMAC, and attaches the HMAC value to the end of the data packet. To ensure that the data cannot be tampered with during transmission, the local controller uses the private key to digitally sign the data packet, and the signature result is attached to the data packet. The receiving end will first verify the digital signature and HMAC before decrypting the data to ensure data integrity and authentication effectiveness.

[0027] S2: The local controller collects the sensor data and transmits it to the edge gateway for a first verification.

[0028] Further, when the device starts up, a shared key encrypted once is obtained through a secure channel; after each data is collected, the AES encryption function is first called to encrypt the data packet, and then the HMAC value is calculated and attached to the end of the data packet.

[0029] The receiving end verifies the HMAC before decrypting to ensure that the data has not been tampered with during transmission.

[0030] The local controller performs preliminary noise reduction and correction on the data; after collecting the data from multiple sensors in the vehicle, the local controller packages it into a standard data packet; the local controller encrypts the data packet once and adds a digital signature, and is ready to upload.

[0031] The edge gateway establishes a connection with each local controller through short-distance communication. After performing identity authentication and handshake to exchange keys, the data packet is uploaded to the edge gateway.

[0032] The edge gateway performs a primary check, including decrypting the received data packet and performing HMAC verification, and then aligning the data using timestamps.

[0033] Data collection and time alignment. Assume that within the time window a dataset from the temperature and humidity sensor, vibration sensor, and GPS module is collected: ; where is the sampling time, is the sensor reading, represents the data type, represents the vehicle or warehouse identifier. For any two data and , if is satisfied, they are grouped into the same time window for fusion. The parameter is the time alignment tolerance.

[0034] For multi-channel data at the same position within the same time window, the weighted average method is used for fusion. Assume that the value collected by the temperature sensor is , and the corresponding weight is (the weight can be set according to the accuracy and historical stability of the sensor), then the fused temperature value can be calculated as: ; Similarly, data such as humidity and vibration are also fused according to a similar formula. At the same time, the edge gateway supplements the GPS positioning data, associates the fused environmental parameters with the corresponding GPS coordinates, and forms a comprehensive data record, which is expressed by the formula: ; where is the fused humidity value, is the fused vibration value, is the fused timestamp. This record is locally stored as a cache backup for subsequent anomaly detection and tracking.

[0035] The edge gateway embeds an adaptive judgment module for local detection. Its main function is to perform anomaly detection on data and adaptively adjust the detection threshold and judgment criteria according to the feedback from the cloud platform.

[0036] It should be noted that the entire data fusion and upload process uses the MQTT protocol, combined with QoS settings and multi-level data verification (including digital signature, HMAC verification, and relay device verification) to ensure the integrity and reliability of data in the transmission link. After being fused and processed by the edge gateway, the data is uploaded to the cloud platform in a standardized message format, providing accurate and real-time data support for subsequent scheduling, path planning, and anomaly warning.

[0037] S3: After the initial verification is completed, local detection is carried out, and the data is fused and uploaded to the cloud platform.

[0038] Furthermore, the local detection includes that when the edge gateway identifies abnormal temperature and humidity, it immediately displays a warning on the local controller of the on-vehicle terminal and sends an adjustment instruction to the local controller to adjust the temperature and humidity; if it cannot be adjusted, the abnormal data is transmitted to the cloud platform, and the cloud platform predicts the temperature control change trend and issues an adjustment instruction, which is fed back to the vehicle terminal via the edge gateway to complete the parameter adjustment; if it still cannot be adjusted, the cloud platform starts to adjust the distribution plan to ensure the safety of the goods.

[0039] Perform initial threshold setting. According to the preset rules, set the initial anomaly thresholds of environmental parameters. Taking temperature, humidity, and vibration as examples, let the initial thresholds be: temperature anomaly threshold , humidity anomaly threshold , vibration anomaly threshold .

[0040] For the fused data , the edge gateway executes the following detection algorithm: If , then it is determined that the temperature is abnormal; if , then it is determined that the humidity is abnormal; if , then it is determined that the vibration is abnormal.

[0041] Among them, , , are the expected values calculated based on historical data (such as using moving average or exponential weighted average), and the detection thresholds 、 、 The initial value is 、 、 , but can be dynamically updated later.

[0042] To achieve intelligent detection, a lightweight reinforcement learning RL algorithm is adopted. Let the state be the state vector composed of the current fusion data and historical anomaly records, and the action be to increase, decrease or keep the current threshold unchanged. Using the Q-learning algorithm, its update formula is expressed as: ; where: is the learning rate; is the discount factor; is the immediate reward after performing the action in the current state, and the reward value can be set according to the anomaly detection accuracy, scheduling feedback effect and operation indicators (such as transportation time consumption, anomaly response rate).

[0043] Meanwhile, the edge gateway updates the detection threshold according to the latest policy parameters (denoted as ) synchronously issued by the cloud platform. The update formula is expressed as: ; ; ; where, is the parameter update factor, reflecting the influence degree of cloud feedback on local adjustment. Historical data and local anomaly events are recorded for subsequent model training and parameter optimization.

[0044] The edge gateway decrypts and verifies the data, and conducts a check to judge the device identity and data integrity.

[0045] Fuse the data within the same time period, the same vehicle and warehouse, supplement the GPS positioning data, associate the temperature, humidity and vibration data with the current location, and store a cache backup locally.

[0046] An adaptive judgment module is embedded in the edge gateway for local detection. The initial state sets the environmental anomaly threshold according to preset rules; the adaptive judgment module records historical data and local anomaly events, and uses a lightweight reinforcement learning algorithm and a parameter adaptive update mechanism that synchronizes with the cloud platform to dynamically adjust the local anomaly detection threshold and judgment criteria according to the latest policy parameters fed back by the cloud.

[0047] First-level response: If the local controller successfully adjusts the temperature and humidity parameters after receiving the instruction and shows a return to the reasonable range in the subsequent data, the local anomaly is considered to have been processed.

[0048] Second-level response: If the local controller cannot adjust the abnormal parameters to the reasonable range, the abnormal data is transmitted to the cloud platform through the edge gateway.

[0049] The cloud platform adopts a temperature control change trend prediction algorithm (such as time series prediction based on the LSTM network), calculates the temperature change trend in the next period of time, and generates a new adjustment instruction: ; The cloud platform compares the prediction result with the current state, generates an instruction and sends it to the edge gateway through the wireless network, and feeds it back to the vehicle terminal for further adjustment.

[0050] For the three-level response, if the parameters cannot be adjusted to the expected range in two consecutive levels of response, the cloud platform will start to adjust the delivery plan. During the adjustment of the delivery plan, the detected delay risk and cargo risk are used as inputs, combined with the vehicle's GPS, RFID, and environmental data, and the optimal delivery route is recalculated through the path planning algorithm, and the driver and the dispatching center are prompted in the system warning. At the same time, the cloud platform plans the nearest distribution center and warehousing center as emergency dispatching nodes to ensure the safety of the goods.

[0051] The adjustment of the delivery plan includes adjusting the travel route when there are delay risks and cargo risks; the GPS, RFID, and environmental sensors in the delivery vehicle collect data in real time to ensure that each package is recorded during the outbound, transportation, and delivery links; the edge gateway integrates the vehicle's driving trajectory and road conditions information and sends it to the cloud platform through the wireless network; the cloud platform analyzes the data in real time and optimizes the delivery route in combination with traffic information.

[0052] When there is a risk of delivery delay, the system automatically gives an early warning and feeds back the new delivery plan to the driver and the dispatching center; when there is a cargo risk, the system automatically gives an early warning and plans the nearest distribution center and warehousing center for emergency dispatching.

[0053] The GPS, RFID, and environmental sensors in the delivery vehicle collect data in real time, and the edge gateway integrates the vehicle's driving trajectory and real-time road conditions information. After data fusion, a comprehensive dynamic data set is formed for the cloud platform to use for real-time analysis.

[0054] The cloud platform analyzes the data in real time and establishes a delay risk index and a cargo risk index mathematical model, which is expressed by the formula: ; ; Among them, is the weight parameter; and are the planned and actual travel times respectively; Tr is the real-time traffic congestion index.

[0055] When or When the preset threshold is exceeded, the system automatically starts path planning. Using a path optimization algorithm, the optimal delivery route is calculated by combining real-time traffic data and historical trip data. For the risk of goods, the system calculates the nearest distribution center or storage center and sends it to the dispatching center as an emergency node.

[0056] The step of fusing and uploading data to the cloud platform includes that the edge gateway prepares the fused data for uploading to the cloud platform; packs the cached data into messages conforming to the MQTT protocol format with QoS settings attached; embeds device identification, timestamp, data type, data content, and checksum information in each message, and then performs encryption processing.

[0057] During the transmission process, the data packet is transmitted to the cloud platform through relay devices. Each relay device performs data verification and then forwards it to ensure data integrity.

[0058] It should be noted that in the adaptive adjustment mechanism based on cloud feedback, by introducing a secure channel, digital signature, HMAC verification, time alignment, weighted data fusion, and RL algorithm to optimize detection parameters online, a closed-loop intelligent monitoring system from sensor data collection to anomaly warning and from local adjustment to global scheduling is realized. This system can continuously adjust the detection threshold and scheduling strategy according to the actual operation feedback, significantly improving the accuracy of anomaly response and the intelligence level of transportation scheduling, and providing a secure, real-time, intelligent, and dynamically adaptive technical solution for the logistics monitoring system.

[0059] S4: The cloud platform conducts a secondary check, detects anomalies in the data and generates early warnings, and sends them to the edge gateway for scheduling.

[0060] Furthermore, the secondary check performed by the cloud platform includes that the cloud receiving module continuously listens for data uploads from the edge gateway; after the cloud platform receives the data, it conducts a secondary check, performs digital signature verification, integrity check, timestamp comparison, and determines whether there are marked abnormal data.

[0061] The data packet format is defined as: ; where is the status code. After the data passes the secondary verification, it is stored in the distributed database and data lake according to the preset classification rules (based on device number, data type, collection period, etc.). At the same time, a log record is generated : ; for subsequent traceability and auditing. Among them, represents the device ID, represents the timestamp, represents the data value, represents the status code, Indicates a digital signature.

[0062] When abnormal data is detected, a warning record is generated and sent to the edge gateway for specific scheduling; and it is pushed to the scheduling center and the on-site monitoring system through the message queue.

[0063] Qualified data is stored in the distributed database and the data lake according to the preset classification rules, and corresponding log records are generated at the same time.

[0064] The abnormal detection includes that after the data is stored in the cloud platform, the background data processing module is immediately started.

[0065] After obtaining the full-link data by using the cloud RL module, the actual scheduling effect and the abnormal handling situation are used as feedback content to calculate rewards and punishments.

[0066] The feedback content includes the actual time consumption of the transportation path, the abnormal warning accuracy rate, and the operation indicators after the execution of the scheduling instruction. A reward function R is constructed to evaluate the current scheduling and abnormal detection strategies, and its form is the weighted sum of the feedback content.

[0067] The scheduling strategy and the abnormal detection standard are updated by using deep reinforcement learning. Let the current state s include the current scheduling strategy, the device state, and the historical abnormal data, and the action a represents the adjustment of the scheduling instruction or the detection parameter. The DQN objective is to minimize the following loss function: ; Where: is the output of the policy network, is the current network parameter; is the target network parameter; is the discount factor; is the immediate reward R; the updated optimal parameter is denoted as .

[0068] The cloud RL module updates the policy online based on the full-link data feedback and outputs the optimized parameters .

[0069] The cloud sends the updated parameter set to each edge gateway through a secure channel (TLS-encrypted MQTT message), and the message format is: ; Where is the update timestamp, and are the digital signature and the HMAC value respectively to ensure message security.

[0070] After the edge gateway receives the updated parameters, it first verifies the message integrity and signature. Let the original local adaptive detection model parameters be , then the updated parameters are smoothly adjusted according to the following formula: ; where is the adjustment rate parameter, which controls the update amplitude of the local model parameters of the edge gateway. The updated parameters are immediately applied to the local detection algorithm, so as to realize the intelligent optimization of the edge layer and the real-time adaptation of the anomaly detection standard.

[0071] It should be noted that based on the feedback results, the cloud RL model continuously updates the online strategy, outputs the optimized scheduling, path and anomaly detection standards in real time, and sends the update results to each edge gateway through a secure channel; after the edge gateway receives the updated parameters, it immediately adjusts the parameters of the local adaptive detection model to realize the intelligent adjustment of the edge layer.

[0072] Embodiment 2, an embodiment of the present invention, provides an Internet of Things-based intelligent logistics monitoring method. In order to verify the beneficial effects of the present invention, scientific demonstrations are carried out through economic benefit calculation and simulation experiments.

[0073] The Internet of Things-based intelligent logistics monitoring system in this embodiment aims to verify the performance of the system in aspects such as data collection, data verification, anomaly detection and scheduling optimization during the transportation process, ensure the stability of the transportation environment, and improve the intelligence level of logistics management.

[0074] Test equipment and environment. In this experiment, five intelligent cold chain transport vehicles (numbered V1, V2, V3, V4, and V5 respectively) are selected for testing, and each vehicle is equipped with the following sensors and computing devices: Temperature and humidity sensor (accuracy: ±0.1°C, ±1%RH), vibration sensor (range: 0-50g, accuracy: 0.01g), GPS positioning module (accuracy: ±2m), RFID scanning module (identification distance: 0-5m), local controller (data preprocessing, encryption module), edge gateway (data fusion and early warning), cloud platform (data storage, analysis and scheduling).

[0075] Before the vehicle starts, all sensors perform self-checks to detect whether the temperature and humidity, vibration, GPS positioning and RFID modules are working properly.

[0076] The local controller establishes a connection and confirms that all devices are online normally. Data encryption and primary verification. After the vehicle runs, the temperature and humidity, vibration, GPS and RFID data are collected at a fixed cycle of 10 seconds.

[0077] The collected data is encrypted by AES and the HMAC value is calculated, and then uploaded to the edge gateway. The edge gateway decrypts the data, verifies the HMAC, and aligns the data through timestamps.

[0078] Local detection and data fusion. All sensor data of the same transport vehicle in the same time period are fused and stored in the cache of the edge gateway.

[0079] Use the adaptive detection module to identify anomalies in data such as temperature, humidity, and vibration. If anomalies are detected, such as the temperature and humidity exceeding the set thresholds, a warning will be immediately displayed on the vehicle terminal, and the abnormal data will be cached.

[0080] Data upload and secondary verification. The edge gateway packages the fused data into messages conforming to the MQTT protocol, and after secondary encryption, uploads them to the cloud platform. The cloud platform conducts secondary verification, including data integrity check, digital signature verification, and timestamp comparison.

[0081] After detecting abnormal data, the cloud platform generates a warning record and issues adjustment instructions to the edge gateway and the dispatching center.

[0082] The cloud platform combines the full-link data and optimizes the dispatching strategy through reinforcement learning (RL). If the risk of delivery delay is detected, the system automatically generates a new delivery plan and notifies the driver to adjust the route.

[0083] It can be seen from the experimental content that the sensor data of all vehicles can be double-checked in the edge gateway and the cloud platform, and the data integrity is effectively guaranteed. In addition, the vibration sensor can accurately identify abnormal shaking conditions during transportation (such as 0.20g of V3) and trigger the corresponding warning function.

[0084] The number of abnormal warnings for vehicles V1, V3, and V5 is slightly higher than that of other vehicles because there are short-term deviations in the temperature and humidity data of these vehicles. For example, the temperature of V3 reaches 5.1°C and the humidity reaches 68%, exceeding the set safety thresholds, resulting in the system triggering a warning. This function ensures the controllability of the transportation environment in the cold chain logistics process and effectively reduces the risks of food and drug transportation.

[0085] Based on the full-link data analysis, the cloud platform can optimize the delivery plan in real time. For example, due to the high vibration value and humidity fluctuation of V3, the system pushed adjustment suggestions, enabling it to adjust the storage method at the next stop. In addition, the RFID recognition rate remains above 97%, demonstrating the high efficiency of the local controller in the data encryption and parsing process.

[0086] Embodiment 3 is an embodiment of the present invention, which provides an intelligent logistics monitoring system based on the Internet of Things, including a sensor module, a local control module, an edge gateway module, and a cloud platform module.

[0087] The sensor module collects temperature, humidity, vibration, GPS, and RFID data after self-check and sends them to the local control.

[0088] The local control module includes a local controller. The local controller performs AES encryption, HMAC calculation, and digital signature on the data. After completing the security encapsulation, it sends device online confirmation and data packets to the edge gateway through a secure channel.

[0089] After receiving the data, the edge gateway module decrypts, verifies, and conducts a primary check; aligns and fuses the data using timestamps, and the local adaptive detection module detects anomalies and gives on-site feedback warnings; packs the data into MQTT messages and uploads them to the cloud platform through a wireless network.

[0090] The cloud platform module is used to perform digital signature verification, integrity check, timestamp comparison, and secondary check after receiving the data packets; stores qualified data in a distributed database, generates warnings for abnormal data and issues scheduling instructions, and at the same time, the cloud RL module performs feedback optimization on the full-link data.

[0091] If the function is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art or a part of this technical solution can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods in various embodiments of the present invention. The aforementioned storage medium includes: USB flash drives, mobile hard disks, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), magnetic disks, or optical discs and other various media that can store program codes.

[0092] The logic and / or steps represented in the flowchart or described in other ways herein, for example, can be considered as a definite sequence list of executable instructions for implementing logical functions, and can be specifically implemented in any computer-readable medium for use by an instruction execution system, apparatus, or device (such as a computer-based system, a system including a processor, or other systems that can fetch instructions from the instruction execution system, apparatus, or device and execute the instructions), or used in combination with these instruction execution systems, apparatus, or devices. For the purposes of this specification, a "computer-readable medium" can be any device that can contain, store, communicate, propagate, or transmit a program for use by or in connection with an instruction execution system, apparatus, or device.

[0093] More specific examples (a non-exhaustive list) of computer-readable media include the following: an electrical connection (electronic device) having one or more wirings, a portable computer diskette (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disc read-only memory (CDROM). Additionally, the computer-readable media can even be paper or other suitable media on which a program can be printed, since the program can be obtained electronically, for example, by optically scanning the paper or other media, followed by editing, interpretation, or, if necessary, other suitable processing, and then storing it in a computer memory.

[0094] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, any one or a combination of the following techniques well known in the art can be used: discrete logic circuits having logic gate circuits for implementing logical functions on data signals, application-specific integrated circuits having appropriate combinational logic gate circuits, programmable gate arrays (PGAs), field programmable gate arrays (FPGAs), etc. It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solutions of the present invention can be modified or equivalently replaced without departing from the spirit and scope of the technical solutions of the present invention, and they should all be covered by the scope of the claims of the present invention.

[0095] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solutions of the present invention can be modified or equivalently replaced without departing from the spirit and scope of the technical solutions of the present invention, and they should all be covered by the scope of the claims of the present invention.

Claims

1. A smart logistics monitoring method based on the Internet of Things, characterized in that, Including: The sensor performs self-check and collects data; The local controller collects the sensor data and transmits it to the edge gateway for a primary verification; After the primary verification is completed, local detection is carried out, the data is fused and uploaded to the cloud platform; The cloud platform conducts a secondary verification, performs anomaly detection on the data and generates a warning, and feeds it back to the edge gateway for scheduling.

2. The intelligent logistics monitoring method based on the Internet of Things according to claim 1, wherein: The self-check of the sensor includes that the sensor includes a temperature and humidity sensor, a vibration sensor, a positioning module and an RFID; before the transport vehicle departs, the sensors in the vehicle perform a self-check to confirm that the temperature, humidity, GPS and RFID devices are operating normally; After the vehicle starts, the sampling period is set using a timer interrupt, and the temperature, humidity, vibration, GPS and RFID data are collected every preset period; the sensor converts the analog data into digital signals and attaches a timestamp and a device number and then transmits them to the local controller.

3. The intelligent logistics monitoring method based on the Internet of Things according to claim 2, wherein: The collection of the sensor data by the local controller includes that when the device starts, the shared key is obtained through a secure channel; after each data collection, the AES encryption function is first called to encrypt the data packet, and then the HMAC value is calculated, and the HMAC is attached to the end of the data packet; The receiving end verifies the HMAC before decryption to ensure that the data has not been tampered with during transmission; The local controller performs preliminary noise reduction and correction on the data; after the data of multiple sensors in the vehicle are collected by the local controller, they are packaged into a standard data packet; The local controller encrypts the data packet and adds a digital signature and prepares to upload it; The edge gateway establishes a connection with each local controller through short-range communication. After authentication and handshake to exchange keys, the data packet is uploaded to the edge gateway.

4. The intelligent logistics monitoring method based on the Internet of Things according to claim 3, wherein: The primary verification by the edge gateway includes that the received data packet is first decrypted and the HMAC is verified, and then the data is aligned using the timestamp; The edge gateway decrypts and verifies the data, conducts a primary verification, and judges the device identity and data integrity; The data in the same time period, the same vehicle and the warehouse are fused, the GPS positioning data is supplemented, the temperature, humidity and vibration data are associated with the current location, and a cache backup is stored locally; An adaptive judgment module is embedded in the edge gateway for local detection, and the environmental anomaly threshold is set according to preset rules in the initial state; The adaptive judgment module records historical data and local anomaly events, and uses a lightweight reinforcement learning algorithm and a parameter adaptive update mechanism that synchronizes with the cloud platform. According to the latest policy parameters fed back by the cloud, the local anomaly detection threshold and judgment criteria are dynamically adjusted; The local detection includes that when the edge gateway identifies abnormal temperature and humidity, a warning is immediately displayed on the local controller of the in-vehicle terminal, and an adjustment instruction is sent to the local controller to adjust the temperature and humidity; If it cannot be adjusted, the abnormal data is transmitted to the cloud platform. The cloud platform predicts the temperature control change trend, issues an adjustment instruction, and feeds it back to the vehicle terminal via the edge gateway to complete the parameter adjustment; if it still cannot be adjusted, the cloud platform starts to adjust the distribution plan; The adjustment of the delivery plan includes adjusting the traveling route when there are delay risks and cargo risks. During the adjustment, the GPS, RFID, and sensors in the delivery vehicle collect multi-source data in real time. The edge gateway integrates the vehicle driving trajectory, road conditions information, and the collected multi-source data, and sends them to the cloud platform through a wireless network. The cloud platform analyzes the uploaded data in real time and optimizes the delivery route in combination with traffic information. Among them, when there is a risk of delivery delay, the cloud platform issues a warning and feeds back the new delivery plan to the driver and the dispatching center. When there is a cargo risk, the cloud platform issues a warning and plans the nearest distribution center and warehousing center for emergency dispatching.

5. The intelligent logistics monitoring method based on the Internet of Things according to claim 4, characterized in that: The process of fusing and uploading data to the cloud platform includes the edge gateway preparing to upload the fused data to the cloud platform. The cached data is packaged into messages conforming to the MQTT protocol format, with QoS settings attached. Device identification, timestamp, data type, data content, and checksum information are embedded in each message, and then encryption processing is performed. The data packet is transmitted to the cloud platform through relay devices during the transmission process, and each relay device forwards the data after data verification.

6. The intelligent logistics monitoring method based on the Internet of Things according to claim 5, characterized in that: The secondary verification performed by the cloud platform includes the cloud receiving module continuously listening for data uploads from the edge gateway. After the cloud platform receives the data, it performs secondary verification, including digital signature verification, integrity check, timestamp comparison, and detection of abnormal data. When abnormal data is detected, a warning record is generated and sent to the edge gateway for specific dispatching, and it is pushed to the dispatching center and in-vehicle terminal through a message queue. After the secondary verification is completed, the uploaded data is stored in a distributed database and a data lake according to preset classification rules, and corresponding log records are generated simultaneously.

7. The intelligent logistics monitoring method based on the Internet of Things according to claim 6, wherein: The abnormal detection process includes that after the data is stored in the database, the background data processing module is immediately started. After the cloud RL module obtains the full-link data, it calculates rewards and punishments with the actual dispatching effect and abnormal handling situation as feedback content. The feedback content includes the actual time consumed for the transportation path, the accuracy rate of abnormal warnings, and the operation indicators after the execution of the dispatching instructions. Based on the feedback results, the cloud RL model continuously updates the strategy online in real time, outputs optimized dispatching, path, and abnormal detection criteria, and sends the update results to each edge gateway through a secure channel. After receiving the updated parameters, the edge gateway immediately adjusts the parameters of the local adaptive detection model.

8. A system adopting the Internet of Things-based intelligent logistics monitoring method as described in any one of claims 1 to 7, characterized in that: It includes a sensor module, a local control module, an edge gateway module, and a cloud platform module. The sensor module, after self-check, collects temperature, humidity, vibration, GPS, and RFID data, and sends them to the local controller. The local control module includes a local controller. The local controller encrypts the data with AES, calculates HMAC, and performs digital signature. After completing the security encapsulation, it sends the device online confirmation and data packet to the edge gateway through a secure channel. After receiving the data, the edge gateway module decrypts, verifies, and performs a primary check. It aligns and fuses the data using timestamps, and the local adaptive detection module detects abnormalities and gives on-site feedback warnings. The data is packaged into MQTT messages and uploaded to the cloud platform through a wireless network. The cloud platform module is used to perform secondary verification after receiving data packets, generate warnings and issue scheduling instructions. At the same time, the cloud RL module optimizes the feedback of the full-link data.

9. A computer device, comprising a memory and a processor, the memory storing a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the Internet of Things-based intelligent logistics monitoring method described in any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the Internet of Things-based intelligent logistics monitoring method described in any one of claims 1 to 7.

Images