Logistics transportation whole-process supervision method and system based on edge computing

Abstract

The present application relates to the technical field of logistics transportation management, and particularly relates to a logistics transportation whole-process supervision method and system based on edge computing. The method comprises the following steps: acquiring real-time dynamic data and static business data of a transportation vehicle; based on the real-time dynamic data, calculating a transportation environment fusion index at the current time; combining the transportation environment fusion index and the cargo sensitivity coefficient to generate a supervision urgency index; in response to the supervision urgency index, and according to the remaining power of the vehicle-mounted edge computing terminal, determining a dynamic computing power adjustment coefficient; based on the dynamic computing power adjustment coefficient, adjusting the sampling frequency of the sensors connected to the edge computing terminal, and / or switching the AI detection model running on the edge computing terminal, so as to realize the supervision of the cargo in the logistics transportation. That is, the scheme of the present application can accurately and reliably perform whole-process supervision of logistics transportation.

Description

Technical Field

[0001] This invention relates to the field of logistics and transportation management technology. More specifically, this invention relates to a method and system for monitoring the entire logistics and transportation process based on edge computing. Background Technology

[0002] With the popularization of IoT technology, the logistics industry is gradually transforming from manual supervision to intelligent supervision. The core technical means of supervision usually involves deploying edge computing terminals on transport vehicles, connecting to multi-source data acquisition devices such as cameras, GPS, accelerometers, and temperature sensors. AI algorithms run locally to monitor driving behavior and cargo status in real time, and upload abnormal results to the cloud to solve the bandwidth bottleneck and high latency problems caused by uploading full data.

[0003] However, existing edge computing monitoring solutions typically employ static resource allocation strategies. This means that regardless of whether a vehicle is traveling at high speed, in congested traffic, or parked, the edge terminal performs continuous monitoring with a fixed sampling frequency and a fixed model (such as always running a computationally expensive deep neural network). This one-size-fits-all approach presents significant technical problems: 1. Waste of computing power and energy: In low-risk scenarios (such as vehicles driving at a constant speed on smooth roads or parked empty), continuous high-frequency detection consumes a large amount of power and computing resources of the vehicle terminal, leading to overheating of the equipment or insufficient power of the vehicle terminal at critical moments.

[0004] 2. Regulatory failure in high-risk scenarios: In extremely high-risk scenarios (such as sharp turns at high speeds or when passing through bumpy roads in rain or snow), a fixed sampling frequency may not be sufficient to capture instantaneous abnormal changes (such as instantaneous displacement of goods or millisecond-level micro-collisions), resulting in the loss of critical regulatory data.

[0005] 3. Lack of scenario adaptability, that is, the system cannot perceive the current logistics context (such as the value of goods, the complexity of road conditions) information, and cannot achieve a dynamic balance between energy saving and high precision.

[0006] Therefore, it is particularly important to adapt to different scenario requirements when supervising the entire logistics and transportation process in order to achieve accurate full-process supervision. Summary of the Invention

[0007] The purpose of this invention is to propose a method and system for monitoring the entire logistics transportation process based on edge computing, in order to solve the problem that the existing technology cannot accurately monitor the entire logistics transportation process; to this end, this invention provides solutions in the following two aspects.

[0008] In a first aspect, the edge computing-based end-to-end logistics transportation monitoring method provided by the present invention includes: Acquire real-time dynamic data and static business data of transport vehicles, wherein the real-time dynamic data includes vehicle motion status information and external environment information, and the static business data includes cargo sensitivity coefficient; Based on the aforementioned real-time dynamic data, a transportation environment fusion index is calculated; By combining the transportation environment fusion index and the cargo sensitivity coefficient, a regulatory urgency index is generated that characterizes the priority of current regulatory needs. In response to the regulatory urgency index and based on the remaining battery power of the vehicle-mounted edge computing terminal, a dynamic computing power adjustment coefficient is determined to adjust the working state of the edge computing terminal. Based on the dynamic computing power adjustment coefficient, the sampling frequency of the sensors connected to the edge computing terminal is adjusted, and / or the corresponding AI model running on the edge computing terminal is switched, so as to realize the supervision of goods in logistics transportation.

[0009] Optionally, the integrated transportation environment indicators include: The instantaneous speed of the transport vehicle, the intensity of vehicle vibration, and the external environmental weather conditions are normalized and then weighted and combined.

[0010] Optionally, the regulatory urgency index The calculation method is as follows: ; in, The transportation environment fusion index at time t; For cargo sensitivity coefficient; This refers to the time interval since the last upload of valid regulatory data; The preset maximum allowed silence time threshold; This is the gain coefficient.

[0011] Optionally, the dynamic computing power adjustment coefficient The calculation method is as follows: ; in, The regulatory urgency index at time t; The remaining battery percentage of the edge computing terminal at time t; This is a power protection factor.

[0012] Optionally, switching the corresponding AI model running on the edge computing terminal includes: When the dynamic computing power adjustment coefficient is lower than the first preset threshold, the system enters a low-power mode, shuts down the AI ​​model, and retains only the threshold determination algorithm based on the IMU sensor. When the dynamic computing power adjustment coefficient is greater than or equal to the first preset threshold and less than the second preset threshold, a lightweight network model is started to perform keyframe analysis on the video stream. When the dynamic computing power adjustment coefficient is greater than or equal to the second preset threshold, the AI ​​model is used to perform concurrent detection on multiple video streams, and a full data caching mechanism is started.

[0013] Optionally, it also includes: after detecting an abnormal event under the selected AI model, locking a data segment containing the time period before and after the abnormal event; determining the upload priority of the data segment based on the dynamic computing power adjustment coefficient; if the dynamic computing power adjustment coefficient is greater than a set threshold, then transmitting alarm information and data segments concurrently through multiple links; if the dynamic computing power adjustment coefficient is less than or equal to the set threshold, then recording a log and uploading it when the network is idle, wherein the set threshold is greater than a second preset threshold.

[0014] Optionally, adjusting the sampling frequency of the sensors connected to the edge computing terminal includes: The sampling frequency of any sensor is set to the sum of the minimum sampling frequency and the corrected sampling frequency, wherein the corrected sampling frequency is the product of the base frame rate and the dynamic computing power adjustment coefficient.

[0015] Optionally, obtaining the real-time dynamic data includes: The vehicle's instantaneous speed and steering angle are obtained via the CAN bus; Vehicle attitude and vibration frequency are acquired via IMU; The system obtains the current location and road network information via GPS, and acquires environmental and weather characteristics via external cameras.

[0016] Optionally, the sensitivity coefficient of the goods is preset according to the type of the acquired goods attributes, wherein the sensitivity coefficient of fragile goods or cold chain goods is higher than that of ordinary department store goods.

[0017] In the second aspect, the edge computing-based end-to-end logistics and transportation monitoring system includes: processor; The memory stores computer instructions for edge computing-based end-to-end logistics and transportation monitoring, which, when executed by the processor, cause the system to perform the aforementioned edge computing-based end-to-end logistics and transportation monitoring method.

[0018] The beneficial effects of this invention are as follows: The solution of this invention introduces a dynamic computing power adjustment coefficient, which can automatically and significantly reduce the sampling rate and shut down high-energy-consuming models in low-risk scenarios, effectively extending the battery life of the vehicle edge terminal and solving the problems of device overheating and crashes caused by high-load operation at all times.

[0019] At the same time, by combining environmental fusion indicators and cargo sensitivity coefficients, the sampling frequency can be automatically increased to several times that of the conventional mode when the vehicle enters a bumpy road or is traveling at high speed. This allows the system to capture instantaneous anomalies (such as minor cargo collisions or sudden steering) that are missed in traditional solutions due to excessively large sampling intervals, thereby improving the completeness of the evidence for determining liability. Attached Figure Description

[0020] Figure 1 This schematically illustrates the steps of the edge computing-based end-to-end logistics transportation monitoring method in this embodiment. Figure 2 A schematic diagram illustrating the generated regulatory urgency index in this embodiment is shown. Figure 3 The diagram illustrates the structural block diagram of the edge computing-based end-to-end logistics transportation monitoring system in this embodiment. Detailed Implementation

[0021] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention.

[0022] This embodiment provides a method for monitoring the entire logistics transportation process based on edge computing. This method is mainly executed by an in-vehicle edge computing terminal and aims to solve the problems of high data transmission latency, uneven distribution of computing resources, and battery range anxiety in the existing monitoring of the entire logistics transportation process, making the monitoring of the entire logistics transportation process more reliable.

[0023] like Figure 1 As shown, this edge computing-based end-to-end logistics transportation monitoring method specifically includes the following steps: Step S1: Obtain real-time dynamic data and static business data of the transport vehicles.

[0024] The vehicle-mounted edge computing node receives real-time dynamic data and static business data from multiple sensors. Specifically, the real-time dynamic data includes vehicle motion status information and external environment information.

[0025] Vehicle motion status information is acquired via the onboard CAN bus (Controller Area Network) and IMU (Inertial Measurement Unit). The CAN bus uses... The frequency of readings is used to measure the vehicle's instantaneous speed, steering angle, and brake pedal travel; the IMU preferably uses a six-axis sensor to collect the vehicle's three-axis acceleration and three-axis angular velocity, thereby calculating the vehicle's vibration frequency and attitude tilt angle.

[0026] External environmental information includes current location, road network information, and weather characteristics. Location and road network information are provided by a GPS / BeiDou dual-mode positioning module combined with offline map data; weather characteristics are obtained by initially extracting image data from external cameras using traditional image processing algorithms (such as dark channel prior algorithms for detecting haze and raindrop detection algorithms), and classifying the extracted features into levels, or by directly accessing data collected from onboard weather sensors.

[0027] The static business data mainly includes the cargo sensitivity coefficient; this cargo sensitivity coefficient is a dimensionless parameter preset based on cargo attributes, used to quantify the lower limit of the cargo's tolerance to the severity of the transportation environment. In this embodiment, the cargo sensitivity coefficient is normalized to... Within the range.

[0028] As a preferred parameter setting: For general merchandise (such as textiles and plastic products), the sensitivity factor for goods is set as follows: ; For precision electronic instruments, the cargo sensitivity coefficient is set to [value missing]. ; For fragile items (such as glass and ceramics) or cold chain pharmaceuticals, the sensitivity factor is set at [value missing]. .

[0029] Understandably, a high sensitivity coefficient for goods means that even minor environmental fluctuations would require high-priority regulatory mechanisms to be triggered.

[0030] Step S2: Based on the real-time dynamic data, calculate the transportation environment fusion index that characterizes the physical risks of the current transportation environment.

[0031] In this embodiment, considering the potential physical pressure that the current external environment may exert on the transportation process, a transportation environment fusion index is constructed based on real-time dynamic data.

[0032] The process for obtaining the integrated transportation environment indicators is as follows: First, the data in the real-time dynamic data are normalized to obtain normalized data.

[0033] In this embodiment, in order to eliminate the influence of data with different dimensions, the instantaneous speed of the vehicle is first... Vehicle body vibration intensity and the weather characteristics of the external environment Perform max-min normalization.

[0034] Secondly, a weighted summation method is used to construct a transportation environment integrated index. .

[0035] As vehicle speed increases, vibration intensifies, and weather conditions worsen (such as slippery roads due to heavy rain), the probability of physical risk of cargo damage increases non-linearly. Therefore, the instantaneous speed of the vehicle at the current moment... Vehicle body vibration acceleration and the weather characteristics of the external environment .

[0036] In this embodiment, the transportation environment integration index The calculation formula is as follows: ; in, Maximum speed or road speed limit; The maximum acceleration threshold that the cargo can withstand; Let be the weather characteristics at time t; and This is the weighting factor.

[0037] Weather characteristics can be divided into multiple levels, such as 0 representing sunny weather and 1 representing extreme weather such as rain or snow.

[0038] In this embodiment, considering that vibration is the most significant factor causing damage to goods during logistics transportation, this embodiment preferably sets... The remaining Weights are assigned to weather features.

[0039] The above formula shows that when the vehicle speed is faster and the road surface is bumpier, the transport environment fusion index increases non-linearly; at the same time, severe weather will directly increase the pressure as an additional bias term; this index reflects the risk level of accidents or cargo damage caused by the physical environment.

[0040] Step S3: Combine the transportation environment fusion index and the cargo sensitivity coefficient to generate a regulatory urgency index that represents the priority of current regulatory needs.

[0041] Since harsh environmental conditions alone do not necessarily equate to high risk (for example, sand and gravel transport is not sensitive to bumps), a cargo sensitivity coefficient and a time dimension need to be introduced to construct a regulatory urgency index. The calculation model is as follows: ; in, The transportation environment fusion index at time t; For cargo sensitivity coefficient; This is the gain coefficient; This refers to the time interval since the last upload of valid regulatory data; This is the preset maximum allowed silent time threshold.

[0042] The preferred range of the above-mentioned gain coefficient is... .

[0043] The exponent term in the above formula The purpose is to demonstrate the "non-linear amplification effect" of cargo vulnerability. For example, when... At that time, for For ordinary goods, the magnification factor is As for Fragile items, magnification surges to This means that for high-value, fragile goods, even a slight increase in the transport environment integration index can lead to a sharp increase in the urgency index, thus forcing the system into a state of high regulation.

[0044] Among them, the time item Its function is to adjust according to the silence time. The increase, even with the integration of transportation environment indicators Very low, urgency index It will also gradually increase. When When this value approaches 2, the regulatory priority is increased, preventing regulatory blind spots caused by edge terminals not uploading data for a long time.

[0045] Figure 2 The paper provides an example of generating a regulatory urgency index. In the high range (t=40-60), the waveform is very sharp, which means that there is an extremely urgent regulatory need at this stage.

[0046] Therefore, the above formula combines cargo sensitivity with environmental indicators through an exponential function, resulting in extremely high regulatory urgency indices for fragile or high-value goods even with slight environmental deterioration. Simultaneously, the introduction of a time interval term ensures that even in prolonged periods of non-regulation, the urgency index will increase linearly over time, even in stable environments, forcibly triggering an inspection and preventing regulatory blind spots.

[0047] Step S4: In response to the regulatory urgency index, and based on the remaining battery power of the vehicle-mounted edge computing terminal, determine a dynamic computing power adjustment coefficient for adjusting the working state of the edge computing terminal.

[0048] This embodiment finds a balance between urgency and battery life by introducing a power protection mechanism.

[0049] Among them, the dynamic computing power adjustment coefficient The calculation method is as follows: ; in, The regulatory urgency index at time t; Let be the current remaining battery percentage of the edge computing terminal at time t, with a value range of . ; The preferred setting for the power protection factor is [value]. .

[0050] The above formula defines the final resource allocation strategy. When the regulatory urgency index... When the power is increased, the dynamic computing power adjustment coefficient is adjusted. The value then increases; however, this upward trend is suppressed by the battery level, when the battery level... When the denominator is low, the denominator increases, forcing a reduction in computing power consumption to maintain the basic survival of the equipment. This logic dynamically balances business needs with hardware constraints.

[0051] For example, in a scenario where the battery is fully charged, when... When fully charged, When the denominator approaches 1, at this point... The system's computing power is entirely determined by the regulatory urgency index, and there are no restrictions on performance release.

[0052] In low-battery scenarios, when Descending to ( When the battery level is low, if the setting is... Then the denominator becomes At this point, even The final dynamic computing power adjustment coefficient remains unchanged. It will also be forcibly reduced to its original value. Therefore, in a power shortage situation, unless there is an extremely urgent circumstance ( (Extremely high), otherwise the system will automatically suppress high computing power operations, thereby greatly extending the standby time of the device.

[0053] Step S5: Based on the dynamic computing power adjustment coefficient, adjust the sampling frequency of the sensors connected to the edge computing terminal, and / or switch the AI ​​model running on the edge computing terminal.

[0054] In this embodiment, the formula for the sensor's sampling frequency is: ; in, Let be the sampling frequency of the sensor at time t. Minimum sampling frequency, Base frame rate.

[0055] In this embodiment, based on the dynamic computing power adjustment coefficient The system divides the numerical value into three working ranges, each corresponding to a different working mode: (1) Low power mode: The dynamic computing power adjustment coefficient is less than the first preset threshold ( ).

[0056] At this point, all power-consuming neural network models are shut down, only the threshold determination algorithm based on the IMU sensor is retained, and the sensor sampling frequency is reduced to the minimum.

[0057] This threshold determination algorithm based on IMU sensors requires very little CPU resources to detect severe collisions or rollovers.

[0058] In this embodiment, near Only basic heart rate data is maintained. The minimum sampling frequency is... The value can be, for example .

[0059] The aforementioned first preset threshold can be the average of the dynamic computing power adjustment coefficients under normal conditions during the logistics transportation process of multiple vehicles in history.

[0060] (2) Keyframe analysis mode: The dynamic computing power adjustment coefficient is greater than or equal to the first preset threshold and less than the second preset threshold. ): In this embodiment, a lightweight network model is activated to perform keyframe analysis on the video stream, and every... Extract one frame of image every few seconds (e.g., 5 seconds) for preliminary detection of cargo displacement or damage.

[0061] Wherein, the sampling frequency at this time Greater than .

[0062] In this embodiment, the lightweight model preferably adopts the MobileNetV3-SSD or YOLOv5-Nano architecture. These models typically have fewer than 2 million parameters, and the inference latency at the edge can be controlled to within 50ms.

[0063] To balance accuracy and speed, the model input image resolution is set to [value missing]. The confidence threshold is set to .

[0064] The aforementioned second preset threshold can be the average of the dynamic computing power adjustment coefficients in cases where goods were damaged during the logistics transportation of multiple vehicles in history.

[0065] (3) Full-scale supervision mode: The dynamic computing power adjustment coefficient is greater than the second preset threshold ( ): In this embodiment, a high-precision AI model is used to perform concurrent detection on multiple video streams, and a full data caching mechanism is activated. The sampling frequency at this time is... Significantly greater than .

[0066] Among them, the high-precision model preferably adopts the YOLOv8-Medium or Faster R-CNN architecture, which can identify minute cargo cracks, leaks, and unauthorized personnel intrusion.

[0067] Furthermore, when concurrently processing video streams from multiple cameras inside the carriage, if an abnormal event (such as cargo tipping over) is detected, the system will immediately lock onto the period preceding the abnormal event. Seconds later The data is processed in seconds, and the upload priority is determined based on a dynamic computing power adjustment coefficient.

[0068] Among them, if If the value exceeds the set threshold, alarm information and high-definition video clips will be transmitted concurrently via multiple links; if... If the value is less than or equal to the set threshold, the log is recorded to local storage and the interrupted transmission will resume when the network is idle.

[0069] The threshold value set above is greater than the second preset threshold value, and the specific value can be set according to the actual situation.

[0070] The multi-link configuration mentioned above can be 4G / 5G + satellite.

[0071] This invention achieves dynamic and elastic allocation of computing resources during logistics supervision by constructing a regulatory urgency index and computing power adjustment model that incorporates multi-dimensional factors such as environment, goods, time, and power consumption. While ensuring the safety of high-value goods, it effectively extends the battery life of edge devices, enabling full-process supervision of logistics transportation.

[0072] This invention also provides a logistics transportation end-to-end monitoring system based on edge computing. For example... Figure 3 As shown, the system includes a processor and a memory. The memory stores computer program instructions, which, when executed by the processor, implement the edge computing-based end-to-end logistics transportation monitoring method according to the present invention.

[0073] The system also includes other components well known to those skilled in the art, such as communication buses and communication interfaces, the settings and functions of which are known in the art and therefore will not be described in detail here.

[0074] In this invention, the aforementioned memory can be any tangible medium containing or storing a program that can be used or combined with an instruction execution system, apparatus, or device. For example, a computer-readable storage medium can be any suitable magnetic or magneto-optical storage medium, such as Resistive Random Access Memory (RRAM), Dynamic Random Access Memory (DRAM), Static Random Access Memory (SRAM), Enhanced Dynamic Random Access Memory (EDRAM), High-Bandwidth Memory (HBM), Hybrid Memory Cube (HMC), etc., or any other medium that can be used to store desired information and can be accessed by an application, module, or both. Any such computer storage medium can be part of a device or accessible to or connected to a device. Any application or module described in this invention can be implemented by computer-readable / executable instructions stored or otherwise maintained on such a computer-readable medium.

[0075] In the description of this specification, "multiple" means at least two, such as two, three or more, etc., unless otherwise expressly and specifically defined.

[0076] While various embodiments of the invention have been shown and described in this specification, it will be apparent to those skilled in the art that such embodiments are provided by way of example only. Many modifications, alterations, and alternatives will occur to those skilled in the art without departing from the spirit and essence of the invention.

Claims

1. A method for monitoring the entire logistics and transportation process based on edge computing, characterized in that, The method includes: Acquire real-time dynamic data and static business data of transport vehicles, wherein the real-time dynamic data includes vehicle motion status information and external environment information, and the static business data includes cargo sensitivity coefficient; Based on the aforementioned real-time dynamic data, a transportation environment fusion index is calculated; By combining the transportation environment fusion index and the cargo sensitivity coefficient, a regulatory urgency index is generated that characterizes the priority of current regulatory needs. In response to the regulatory urgency index and based on the remaining battery power of the vehicle-mounted edge computing terminal, a dynamic computing power adjustment coefficient is determined to adjust the working state of the edge computing terminal. Based on the dynamic computing power adjustment coefficient, the sampling frequency of the sensors connected to the edge computing terminal is adjusted, and / or the corresponding AI model running on the edge computing terminal is switched, so as to realize the supervision of goods in logistics transportation.

2. The method for monitoring the entire logistics transportation process based on edge computing according to claim 1, characterized in that, The integrated transportation environment indicators include: The instantaneous speed of the transport vehicle, the intensity of vehicle vibration, and the external environmental weather conditions are normalized and then weighted and combined.

3. The method for monitoring the entire logistics transportation process based on edge computing according to claim 1, characterized in that, The regulatory urgency index The calculation method is as follows: ； in, The transportation environment fusion index at time t; For cargo sensitivity coefficient; This refers to the time interval since the last upload of valid regulatory data; The preset maximum allowed silence time threshold; This is the gain coefficient.

4. The method for monitoring the entire logistics transportation process based on edge computing according to claim 1, characterized in that, The dynamic computing power adjustment coefficient The calculation method is as follows: ； in, The regulatory urgency index at time t; The remaining battery percentage of the edge computing terminal at time t; This is a power protection factor.

5. The method for monitoring the entire logistics transportation process based on edge computing according to claim 1, characterized in that, Switching the corresponding AI model running on the edge computing terminal includes: When the dynamic computing power adjustment coefficient is lower than the first preset threshold, the system enters a low-power mode, shuts down the AI ​​model, and retains only the threshold determination algorithm based on the IMU sensor. When the dynamic computing power adjustment coefficient is greater than or equal to the first preset threshold and less than the second preset threshold, a lightweight network model is started to perform keyframe analysis on the video stream. When the dynamic computing power adjustment coefficient is greater than or equal to the second preset threshold, the AI ​​model is used to perform concurrent detection on multiple video streams, and a full data caching mechanism is started.

6. The method for monitoring the entire logistics transportation process based on edge computing according to claim 5, characterized in that, Also includes: After an abnormal event is detected using the selected AI model, the data segment containing the time periods before and after the abnormal event is locked. The upload priority of the data segment is determined based on the dynamic computing power adjustment coefficient; if the dynamic computing power adjustment coefficient is higher than the set threshold, alarm information and data segments are transmitted concurrently through multiple links; if the dynamic computing power adjustment coefficient is lower than the set threshold, a log is recorded and uploaded when the network is idle, and the set threshold is greater than the second preset threshold.

7. The method for monitoring the entire logistics transportation process based on edge computing according to claim 1, characterized in that, Adjusting the sampling frequency of the sensors connected to the edge computing terminal includes: The sampling frequency of any sensor is set to the sum of the minimum sampling frequency and the corrected sampling frequency, wherein the corrected sampling frequency is the product of the base frame rate and the dynamic computing power adjustment coefficient.

8. The method for monitoring the entire logistics transportation process based on edge computing according to claim 1, characterized in that, Obtaining the real-time dynamic data includes: The vehicle's instantaneous speed and steering angle are obtained via the CAN bus; Vehicle attitude and vibration frequency are acquired via IMU; The system obtains the current location and road network information via GPS, and acquires environmental and weather characteristics via external cameras.

9. The method for monitoring the entire logistics transportation process based on edge computing according to claim 3, characterized in that, The sensitivity coefficient of the goods is preset based on the type of the acquired goods attributes, wherein the sensitivity coefficient of fragile goods or cold chain goods is higher than that of ordinary department store goods.

10. A logistics transportation end-to-end monitoring system based on edge computing, characterized in that: include: processor; A memory storing computer instructions for edge computing-based end-to-end logistics transportation monitoring, which, when executed by the processor, cause the system to perform the edge computing-based end-to-end logistics transportation monitoring method according to any one of claims 1-9.

Images