Cold-chain logistics vehicle path planning system based on data analysis

Abstract

The invention discloses a cold-chain logistics vehicle path planning system based on data analysis, and relates to the technical field of cold-chain vehicle path planning. Comprising a data acquisition and sensor scanning module, a data preprocessing and feature extraction module, an environment complexity quantitative analysis module, an intelligent evaluation and classification module, a stability path planning module and a dynamic path planning and real-time adjustment module, the road condition sensing system scans the surrounding environment regularly through a sensor on a vehicle at a preset regular data updating frequency to obtain real-time road traffic images and environment data information, and stable data support is provided for subsequent path planning and driving decision making. By dynamically adjusting the data updating frequency, responding to the road change in real time and optimizing the path planning, the cold-chain logistics efficiency is improved, the aging loss is reduced, and the cargo quality and the customer experience are ensured.

Description

Technical Field

[0001] The present invention relates to the technical field of cold chain vehicle route planning, and particularly to a cold chain logistics vehicle route planning system based on data analysis. Background Art

[0002] Cold chain logistics vehicle route planning based on data analysis is to optimize the route selection of cold chain logistics vehicles by collecting and analyzing a large amount of data related to cold chain transportation (such as temperature, humidity, road conditions, traffic flow, cargo characteristics, delivery timeliness, etc.), and using data analysis technologies and algorithms (such as machine learning, optimization algorithms, etc.). Its goal is to make dynamic adjustments according to possible changes during transportation (such as road congestion, weather changes, vehicle failures, etc.) to ensure that the goods always remain in a suitable temperature-controlled environment during transportation, and to shorten the transportation time as much as possible, reduce costs, and improve transportation efficiency. At the same time, route planning also needs to consider various factors, including delivery requirements, vehicle capacity, resource constraints, etc., so as to maximize the operating efficiency of the logistics system while meeting customer needs.

[0003] The cold chain logistics vehicle route planning based on data analysis will use a road condition perception system. The road condition perception system monitors the road environment around the vehicle by integrating various sensor technologies, such as lidar (LiDAR), millimeter-wave radar, cameras, ultrasonic sensors, etc. Its main function is to detect and analyze obstacles, traffic conditions, road signs, traffic lights, construction areas, accident sections, etc. on the road. These perception information can help cold chain logistics vehicles evaluate the safety and traffic conditions of the current driving environment in real time, and timely identify factors that may affect the transportation process, such as traffic congestion, road damage, construction closures, etc. In addition, the road condition perception system can also communicate with other vehicles and infrastructure to obtain more traffic dynamic information, so as to provide more accurate route planning suggestions for the vehicle. Through the comprehensive perception of the environment, the road condition perception system can automatically adjust the vehicle driving strategy under dynamically changing road conditions, avoid unnecessary stops or detours, ensure that cold chain logistics vehicles can complete transportation tasks efficiently and safely, while maintaining the temperature control requirements of the goods, reducing risks and delays.

[0004] The prior art has the following deficiencies:

[0005] In the prior art, the road condition perception system usually obtains road traffic images and environmental information through regular data updates, and uses sensors on the vehicle (such as cameras, radars, lidar, etc.) to scan the surrounding environment regularly. However, when cold chain logistics vehicles are driving in construction areas or road renovation areas, the uncertainty of the traffic environment greatly increases the complexity of route planning. If still relying on the acquisition method of regular data updates, especially in these dynamically changing areas, it may bring serious consequences. Regular data updates make the system unable to capture the rapid changes in road conditions in real time, especially in high-risk areas such as construction areas and accident-prone areas. When sudden road blockages, construction vehicles or obstacles appear, the system may not be able to refresh the data in time and make corresponding adjustments, resulting in the vehicle continuing to drive on the congested section, failing to quickly optimize the route or adjust the driving strategy. This lagged response will not only significantly delay the transportation time, but also may miss the best detour opportunity, causing serious losses in the timeliness of cold chain logistics, further affecting the customer experience and the quality of goods.

[0006] The above information disclosed in the background art section is only used to enhance the understanding of the background of the present disclosure, and thus it may include information that does not constitute the prior art known to those of ordinary skill in the art. Summary of the Invention

[0007] The object of the present invention is to provide a cold chain logistics vehicle route planning system based on data analysis, which solves the lag problem caused by the fixed acquisition frequency of the traditional road condition perception system by dynamically adjusting the data update frequency. The system captures the surrounding environment in real time through sensors, accurately extracts key features, and dynamically adjusts the data update frequency according to the environmental complexity. In high-complexity areas (such as construction areas), it can quickly respond to road changes and optimize the driving route. In low-complexity environments, it maintains regular updates to ensure stable operation and reduce resource consumption. This flexible strategy improves the route planning efficiency, reduces the timeliness loss, ensures the punctuality of cold chain logistics and the quality of goods, enhances the customer experience and optimizes the logistics efficiency, so as to solve the problems in the above background art.

[0008] To achieve the above object, the present invention provides the following technical solution: A cold chain logistics vehicle route planning system based on data analysis, comprising a data acquisition and sensor scanning module, a data preprocessing and feature extraction module, an environmental complexity quantification and analysis module, an intelligent evaluation and classification module, a stability route planning module, and a dynamic route planning and real-time adjustment module:

[0009] The data acquisition and sensor scanning module, the road condition perception system first scans the surrounding environment regularly through the sensors on the vehicle at a preset regular data update frequency to obtain real-time road traffic images and environmental data information, providing stable data support for subsequent route planning and driving decisions;

[0010] The data preprocessing and feature extraction module preprocesses the acquired data after obtaining the original road traffic images and environmental data information, and extracts key features from the preprocessed data that reflect the current vehicle being in a high-complexity environment;

[0011] The environmental complexity quantification and analysis module deeply analyzes the extracted key features within the detection window, and quantifies the complexity of the vehicle driving environment through the analyzed features;

[0012] The intelligent evaluation and classification module inputs the quantified complexity features into a pre-trained support vector machine model for intelligent evaluation, and classifies the vehicle driving environment into a high-complexity environment and a low-complexity environment;

[0013] The stability path planning module maintains a preset data update frequency in a low-complexity environment to ensure the stability and efficiency of path planning;

[0014] The dynamic path planning and real-time adjustment module dynamically increases the data update frequency in a high-complexity environment, responds in a timely manner to the rapid changes in road conditions, and recalculates and optimizes the vehicle's driving route.

[0015] Preferably, key features that reflect the current vehicle being in a high-complexity environment are extracted from the preprocessed data. The extracted features include the frequency and severity of path blockages, as well as the number and density of informal traffic signs and temporary signs that appear. Within the detection window, after analyzing the frequency and severity of the extracted path blockages, and the number and density of informal traffic signs and temporary signs that appear, a dynamic path blockage reference value and a non-identifying traffic sign reference value are generated. The complexity of the vehicle driving environment is quantified through the dynamic path blockage reference value and the non-identifying traffic sign reference value.

[0016] Preferably, the specific steps for generating a dynamic path blockage reference value by analyzing the frequency and severity of path blockages within the detection window are as follows:

[0017] First, define a path blockage event. To quantify the frequency, set the blockage event incidence rate, which represents the number of blockage events per unit time. The formula for the blockage event incidence rate is as follows:

[0018]

[0019] , where B freq is the path blockage frequency, n is the total number of blockage events that occur within the detection window, D i is the duration of the i-th path blockage event, and P i is the blockage type of the i-th path blockage event. It is an indicator function used to mark whether the path block event i is a valid path block;

[0020] To further improve the quantization accuracy of path blockage, identify the blockage severity, and calculate the blockage impact factor. The calculation formula is as follows:

[0021]

[0022] , where, I block is the blockage impact factor, A j is the area of the j-th blockage area, V j is the traffic flow deceleration factor of the j-th blockage area, P j is the path blockage status of the j-th blockage area, It is an indicator function used to mark whether there is a valid blockage event in this area;

[0023] The comprehensive path blockage frequency B freq and the blockage impact factor I block are used to generate a dynamic path blockage reference value, which is used to represent the complexity of the vehicle's environment. The generation formula is as follows:

[0024]

[0025] , where, DPBF is the dynamic path blockage reference value, I block，k is the path blockage frequency in the k-th time period, I block，k is the path blockage impact factor in the k-th time period, p is the total number of time periods within the monitoring window, and θ is the blockage impact sensitivity adjustment parameter, is the exponential weighting of the blockage severity.

[0026] Preferably, the specific steps for analyzing the number and density of informal traffic signs and temporary signs that appear under the detection window to generate a non-identifying traffic sign reference value are as follows:

[0027] First, collect the number and density data of all informal traffic signs and temporary signs. Since the distribution densities of informal traffic signs and temporary signs in different areas are different, a local density function is needed to analyze the density. The local density function formula is as follows:

[0028]

[0029] , where, ρ(x, y) is the local density function, representing the density of informal traffic signs in the area around the position (x, y), N(x, y) is the number of informal traffic signs around the position (x, y), and A(x, y) is the area of the area at the position (x, y);

[0030] Combined with the local density function ρ(x, y) and the total number of signs, calculate the distribution factor of non - identification traffic signs. The calculation formula is as follows:

[0031]

[0032] , where DI is the distribution factor of non - identification traffic signs, ρ(x q , y q ) is the local density of the q - th sign, N is the total number of informal traffic signs, A is the total area of the region, d(x q ) is the distance from the q - th sign to the center of the region, d max is the maximum distance of the region, representing the maximum distance between any sign in the region and the center of the region;

[0033] After obtaining the distribution factor DI of non - identification traffic signs, further quantify the potential impact of signs on the vehicle driving path through the non - identification traffic sign influence coefficient. The calculation formula of the non - identification traffic sign influence coefficient is as follows:

[0034]

[0035] , where β sign is the non - iconic traffic sign influence coefficient, w q is the weight coefficient of the q - th sign;

[0036] Integrate the distribution factor DI of non - identification traffic signs and the non - identification traffic sign influence coefficient β sign to generate the non - identification traffic sign reference value. The generation formula is as follows:

[0037]

[0038] , where R sign is the non - iconic traffic sign reference value, D max is the theoretical maximum value of the non - identification traffic sign distribution density, β max is the theoretical maximum value of the non - identification traffic sign's influence on the driving path.

[0039] Preferably, input the quantified dynamic path blockage reference value and the non - identification traffic sign reference value into a pre - trained support vector machine model, and generate an environmental complexity index through the support vector machine model, and intelligently evaluate the vehicle driving environment through the environmental complexity index.

[0040] Preferably, compare and analyze the environmental complexity index generated when the vehicle driving environment is intelligently evaluated by a pre - trained support vector machine model with a pre - set environmental complexity index reference threshold, and divide the vehicle driving environment. The division steps are as follows:

[0041] If the environmental complexity index is greater than the pre-set reference threshold of the environmental complexity index, the vehicle driving environment is classified as a high-complexity environment;

[0042] If the environmental complexity index is less than or equal to the pre-set reference threshold of the environmental complexity index, the vehicle driving environment is classified as a low-complexity environment.

[0043] Preferably, in a high-complexity environment, the specific steps to dynamically increase the data update frequency, respond in a timely manner to the rapid changes in road conditions, and recalculate and optimize the vehicle's driving route are as follows:

[0044] When the environmental complexity index ENVC is greater than the reference threshold of the environmental complexity index, it indicates that the environmental complexity where the vehicle is located has exceeded the normal range, and it is necessary to dynamically adjust the data update frequency to respond in a timely manner to the rapidly changing road conditions. At this time, based on the environmental complexity index ENVC and the pre-set data update frequency, calculate the new data update frequency, and the calculation formula is as follows:

[0045]

[0046] , where f0 is the pre-set data update frequency, ENVC is the environmental complexity index, ENVC ref is the reference threshold of the environmental complexity index, α is an adjustment factor used to control the sensitivity of the update frequency, f new is the new data update frequency;

[0047] After adjusting the update frequency, the road condition perception system starts to collect the latest road traffic images and environmental data at the new frequency. At this time, the sensor updates the data according to the adjusted frequency, and through real-time transmission to the central processing system for analysis, the latest data obtained is represented by the following formula:

[0048]

[0049] , where D current (t) is the complete environmental data set at time t, including all sensor data, S p (t) is the raw data provided by the p-th sensor at time t, W p is the weighting coefficient of the p-th sensor data, and v is the total number of sensors;

[0050] After obtaining the latest environmental data, recalculate and optimize the vehicle driving route through the path planning algorithm, and select the optimal path to cope with the dynamically changing road conditions.

[0051] In the above technical solution, the technical effects and advantages provided by the present invention:

[0052] By adjusting the data update frequency in real time, the present invention can effectively solve the problem of reaction lag caused by the fixed data collection frequency in the traditional road condition perception system. In this solution, the system first captures the traffic images and data information of the surrounding environment through sensors, and after preprocessing and key feature extraction, accurately identifies the complexity of the environment where the vehicle is located. In a high-complexity environment, such as a construction area or a road renovation area, the system will dynamically increase the data update frequency, so as to respond in a timely manner to the rapid changes in road conditions, ensure that sudden events (such as traffic jams, obstacles or temporary closures, etc.) can be captured in time, and quickly adjust the driving route to avoid delays and unnecessary detours. On the contrary, in a low-complexity environment, maintaining regular data updates can not only ensure the stable operation of the system, but also reduce resource consumption. This flexible data update strategy not only improves the efficiency and accuracy of path planning, but also effectively reduces the loss of timeliness, ensures the punctuality and cargo quality of cold chain logistics transportation, and improves the customer experience. At the same time, the intelligent evaluation system enhances the rapid response ability to environmental changes, improves the ability to cope with complex traffic situations and uncertain factors, and optimizes the overall efficiency of the logistics process. Brief Description of the Drawings

[0053] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings required to be used in the embodiments. Obviously, the drawings described below are only some embodiments recorded in the present invention. For those of ordinary skill in the art, other drawings can also be obtained according to these drawings.

[0054] Figure 1 It is a schematic diagram of the modules of a cold chain logistics vehicle path planning system based on data analysis according to the present invention. Detailed Embodiments

[0055] Now, the exemplary embodiments will be described more fully with reference to the accompanying drawings. However, the exemplary embodiments can be implemented in various forms and should not be construed as limited to the examples set forth herein; rather, these exemplary embodiments are provided so that the present disclosure will be more complete and comprehensive, and will fully convey the concept of the exemplary embodiments to those skilled in the art.

[0056] The present invention provides a cold chain logistics vehicle path planning system based on data analysis as Figure 1 shown, including a data collection and sensor scanning module, a data preprocessing and feature extraction module, an environmental complexity quantification and analysis module, an intelligent evaluation and classification module, a stability path planning module, and a dynamic path planning and real-time adjustment module:

[0057] Data acquisition and sensor scanning module. First, the road condition perception system regularly scans the surrounding environment through sensors on the vehicle (such as cameras, radars, lidars, etc.) at a preset regular data update frequency to obtain real-time road traffic images and environmental data information, providing stable data support for subsequent path planning and driving decisions;

[0058] The preset regular data update frequency refers to the frequency at which the road condition perception system obtains road traffic images and environmental data information from sensors on the vehicle (such as cameras, radars, lidars, etc.) at a preset fixed time interval. This update frequency is usually preset according to typical traffic conditions and vehicle speeds during system design. For example, scanning once per second or collecting data every 500 milliseconds. This fixed collection interval ensures that the system can continuously and stably obtain road environment information, providing continuous perception data for path planning and driving decisions. However, the preset update frequency is usually static and cannot be adjusted in real time according to the dynamic changes of the traffic environment. Therefore, when the environment undergoes sudden changes or the complexity increases, the fixed update frequency may not be able to capture key change information in time, resulting in a lag in the system's response, thus affecting the accuracy and timeliness of path planning. Therefore, the regular data update frequency is the basis of the road condition perception system but needs to be adjusted in a dynamic environment to adapt to actual traffic changes.

[0059] These sensors can capture key data such as traffic flow, road condition changes, obstacle positions, traffic sign status, and signal light changes. The core of this step is to ensure that the system can continuously and stably collect necessary traffic and environmental information, providing a reliable data basis for subsequent path planning and real-time response. The selection of the preset frequency needs to balance the real-time nature of the data and the computational load of the system to ensure the efficiency and continuity of data collection without excessive consumption of computational resources.

[0060] Data preprocessing and feature extraction module. After obtaining the original road traffic images and environmental data information, the acquired data is preprocessed, and key features reflecting that the current vehicle is in a high-complexity environment are extracted from the preprocessed data;

[0061] The preprocessing process includes data cleaning, denoising, standardization, and format conversion, etc., to ensure data quality and consistency. Subsequently, key features reflecting that the current vehicle is in a high-complexity environment are extracted from the preprocessed data. These key features may include traffic flow density, road closure in construction areas, accident frequency, road narrowness, the number of pedestrians and non-motor vehicles, etc. Through feature extraction, complex raw data can be transformed into meaningful indicators, providing necessary information for subsequent complexity quantification and the input of the support vector machine model.

[0062] The purpose of preprocessing the acquired road traffic images and environmental data information is to improve the quality and usability of the data, reduce the interference of noise and redundant information, and thus provide accurate and reliable basic data for subsequent feature extraction and intelligent evaluation. Since the raw data collected by sensors may contain a large amount of incomplete, repetitive, or interfered information (such as image noise caused by factors like light changes, weather effects, motion blur, etc.), directly using the unprocessed data may lead to inaccurate evaluation results of the system. Therefore, through preprocessing, the data can be denoised, corrected, normalized, and format-converted, eliminating irrelevant or incorrect data, complementing missing information, and making the data more consistent and standardized. Preprocessing also includes enhancing the images, such as edge detection, color correction, resolution adjustment, etc., to ensure that the key features are clearer. The preprocessed data can not only reduce the computational burden of the system but also improve the performance of deep learning models or other analysis algorithms, ensuring that the path planning and decision-making of the system in complex environments are more accurate and efficient.

[0063] The environmental complexity quantification analysis module deeply analyzes the extracted key features within the detection window and quantifies the complexity of the vehicle driving environment through the analyzed features;

[0064] Extract the key features from the preprocessed data that reflect the current vehicle being in a high-complexity environment. The extracted features include the frequency and severity of blocked paths, as well as the number and density of informal traffic signs and temporary signs that appear. Analyze the frequency and severity of blocked paths, the number and density of informal traffic signs and temporary signs that appear under the detection window to generate a dynamic path blockage reference value and a non-identifying traffic sign reference value, and quantify the complexity of the vehicle driving environment through the dynamic path blockage reference value and the non-identifying traffic sign reference value.

[0065] When the frequency of temporary path blockage increases and the blockage time lengthens, it clearly indicates that the vehicle's environment is a high-complexity environment. This is because the dynamic changes in path blockage directly increase the uncertainty of vehicle passage and the complexity of driving decisions. In construction areas or road renovation areas, temporary blockages are usually caused by the movement of construction vehicles, the setting up of roadblocks, or the operations of construction workers. Such blockage events are often sudden, unpredictable, and constantly changing in location and time. If the blockage frequency is high, it means that the vehicle has to stop, decelerate, or detour frequently, and each time it needs to readjust its driving strategy, increasing the computational burden on the driving system. The longer the blockage time, the longer the duration of the dynamic changes in the environment, resulting in a significant decrease in path availability and further increasing the difficulty of path planning. In addition, the frequency and duration of blockage events will trigger chain reactions such as road congestion and traffic flow fluctuations, making it difficult for the system to perform effective path optimization based on static data. High-frequency and long-duration blockage situations make it impossible for the vehicle to drive along the original planned route and must rely on dynamically adjusting the path and real-time sensing of environmental changes. The uncertainty and complexity of this environment are significantly increased. Therefore, the higher the frequency and duration of path blockage, the more it indicates that the vehicle is in a high-complexity environment, and the system needs to increase the data update frequency and the flexibility of path planning in real time.

[0066] The specific steps for analyzing the frequency and severity of path blockage under the detection window to generate a dynamic path blockage reference value are as follows:

[0067] First, define a path blockage event. A path blockage event refers to any factor that temporarily prevents or restricts vehicle passage, such as construction vehicles, roadblocks, accident scenes, etc. Whenever a path blockage event occurs, record information such as the duration of the event, the number of lanes affected, and the degree of blockage. To quantify the frequency, set the blockage event incidence rate, which represents the number of blockage events per unit time. The formula for the blockage event incidence rate is as follows:

[0068]

[0069] where, B freq is the path blockage frequency, n is the total number of blockage events that occur within the detection window, D i is the duration of the i-th path blockage event, indicating the length of time that the blockage event lasts, P i is the blockage type of the i-th path blockage event, which can be a blockage caused by construction, accident, road closure, etc. This parameter is used to identify whether the event constitutes an effective path blockage. is an indicator function used to mark whether the path blockage event i is an effective path blockage. If P i represents an effective blockage event (for example, the road is blocked due to construction), then (P i ) = 1, otherwise it is 0;

[0070] The goal of this step is to record and quantify the frequency of effective obstacle events occurring within a unit of time. By calculating these events, it preliminarily reflects the dynamic obstacles faced by the vehicle.

[0071] To further improve the quantification accuracy of path blockage and identify the severity of blockage, the severity of blockage depends on its impact on traffic flow, including the scope of path blockage, duration, and the impact on traffic flow in the blocked area, etc. Calculate the blockage impact factor, which represents the impact intensity of path blockage on the overall traffic flow. The calculation formula is as follows:

[0072]

[0073] , where, I block is the blockage impact factor, A j is the area of the j-th blocked area, representing the physical scope of this area, V j is the traffic flow deceleration factor of the j-th blocked area, representing the degree of traffic flow obstruction in this area, P j is the path blockage status of the j-th blocked area, is the indicator function, used to mark whether there is an effective blockage event in this area. If there is a blockage event in this area, then otherwise it is 0;

[0074] Through this step, we can calculate the impact intensity of each blockage event, and then calculate the severity of path blockage. If multiple lanes or roads are blocked, or the traffic flow is significantly reduced, then the severity of this blockage event will be higher.

[0075] Combined with the path blockage frequency B freq and the blockage impact factor I block , generate a dynamic path blockage reference value, which is used to represent the complexity of the vehicle's environment. The generation formula is as follows:

[0076]

[0077] , where, DPBF is the dynamic path blockage reference value, I block，k is the path blockage frequency in the k-th time period, representing the number of blockage events occurring within a unit of time, and is used to measure the dynamic change degree of the path after weighting, I block，k is the path blockage impact factor in the k-th time period, which comprehensively considers factors such as the duration of the blockage event, the impact scope (such as the number of blocked lanes or the area of the region), and the change in traffic flow. p is the total number of time periods within the monitoring window, and θ is the blockage impact sensitivity adjustment parameter, which controls the sensitivity of the dynamic path blockage reference value to the path blockage frequency I block，k . It is the exponential weighting of the blocking severity. The blocking severity of the k-th time period is exponentially weighted, so that the impact of the severity on the dynamic path blocking factor decreases nonlinearly.

[0078] By introducing the weighted index The sensitivity of the factor to severe blocking events can be adjusted dynamically. The purpose of this weighting method is to make the impact of severe blocking events on the dynamic path blocking reference value more significant, while taking into account the time factor.

[0079] The larger the performance value of the dynamic path blocking reference value generated by analyzing the frequency and severity of the path blocking under the detection window, the more frequent the dynamic changes in the vehicle's environment and the higher the traffic uncertainty, which belongs to a high-complexity environment. The higher the performance value, the more temporary blocking events the vehicle encounters in a short period of time, and the longer the duration and the larger the range of the blocking, which greatly reduces the availability of the passage path. The system must frequently adjust the path and optimize the decision to ensure the safety and timely passage of the vehicle. In this case, the system needs to dynamically increase the data update frequency and environmental perception capabilities to cope with changes in the complex environment. On the contrary, if the performance value of the dynamic path blocking reference value is small, it means that the vehicle encounters fewer blocking events on the current path, the blocking time is short and the impact range is small, indicating that the environment in which the vehicle is located is relatively stable and the traffic is smooth, which belongs to a low-complexity environment. The system can maintain the preset data update frequency to ensure the efficiency and stability of path planning.

[0080] When the number of informal traffic signs and temporary signs increases and their distribution density increases, it indicates that the vehicle is in a highly complex environment. This is because informal signs and temporary signs are usually set up in temporary situations such as road reconstruction, construction or traffic control. Their appearance reflects the rapid changes and uncertainty of the road environment. The information of informal signs and temporary signs often does not have a unified standard. They may be composed of handwritten signs, temporarily erected obstacles or construction workers' instructions. Such signs not only do not have standard visual recognition features, but their location, content and effectiveness will change with the progress of construction, greatly increasing the recognition difficulty of drivers or autonomous driving systems. This change makes the environment full of uncertainty, and different types of signs may have conflicting instructions, such as speed limit signs, detour instructions or temporary parking requirements. When the number and density of these signs and signs increase, it means that the traffic rules and driving paths on the road are constantly changing, driving decisions need to be constantly adjusted, and the system must process a large amount of information in real time to ensure that the vehicle passes through the complex environment safely and quickly. Therefore, the increase and dense distribution of informal signs and temporary signs directly reflect the high complexity of road conditions. The frequency and unpredictability of changes in the vehicle environment have also increased significantly, which undoubtedly increases the difficulty of path planning and decision-making, indicating that this is a highly complex environment.

[0081] The specific steps for analyzing the quantity and density of informal traffic signs and temporary signs that will appear under the detection window to generate a reference value for non-identifying traffic signs are as follows:

[0082] First, collect the quantity and density data of all informal traffic signs and temporary signs. Informal traffic signs include all signs that do not meet the specifications, such as handwritten road signs, temporarily erected signs, etc. Temporary signs refer to any temporarily set traffic guiding signs, such as construction fences, temporary speed limit signs, etc. The distribution densities of informal traffic signs and temporary signs are different in different areas, and a local density function is needed to analyze the density. The formula for the local density function is as follows:

[0083]

[0084] , where ρ(x, y) is the local density function, representing the density of informal traffic signs in the area around the position (x, y), N(x, y) is the number of informal traffic signs around the position (x, y), and A(x, y) is the area of the area at the position (x, y);

[0085] Through this step, the local density of each position can be obtained, thereby reflecting the distribution of signs and the complexity of the area.

[0086] Combining the local density function ρ(x, y) and the total number of signs, calculate the distribution factor of non-identifying traffic signs. The distribution factor of non-identifying traffic signs comprehensively considers the combined influence of the number and density of signs and is used to quantify the distribution of signs in the construction area. The calculation expression is as follows:

[0087]

[0088] , where DI is the distribution factor of non-identifying traffic signs, ρ(x q , y q ) is the local density of the qth sign, representing the number of informal traffic signs existing per unit area around the qth sign (x, y), N is the total number of informal traffic signs, A is the total area of the area, d(x q ) is the distance from the qth sign to the center of the area, and d max is the maximum distance of the area, representing the maximum distance between any sign in the area and the center of the area;

[0089] This step comprehensively considers the distribution density of signs and the relative positions of signs, taking into account that the closer a sign is to the center of the area, the greater its impact on the environmental complexity. Therefore, the larger the distribution index DI, the denser the traffic signs in the construction area and the higher the environmental complexity.

[0090] After obtaining the distribution factor DI of non - identifying traffic signs, the potential impact of signs on vehicle driving paths is further quantified by the non - identifying traffic sign influence coefficient. The calculation of this coefficient takes into account not only the distribution index but also factors such as the type and visibility of the signs. The formula for calculating the non - identifying traffic sign influence coefficient is as follows:

[0091]

[0092] , where β sign is the non - identifying traffic sign influence coefficient, w q is the weight coefficient of the q - th sign, reflecting the relative importance of different non - identifying traffic signs to the environmental complexity;

[0093] By combining the weight of the sign with its density influence in the construction area, the contribution of the sign to the driving path complexity is calculated. When β sign is relatively high, it indicates that the traffic signs in the construction area have a greater impact on driving decisions, and vehicles may need to frequently adjust their paths or slow down.

[0094] By synthesizing the distribution factor DI of non - identifying traffic signs and the non - identifying traffic sign influence coefficient β sign a non - identifying traffic sign reference value is generated. The generation formula is as follows:

[0095]

[0096] , where R sign is the non - identifying traffic sign reference value, D max is the theoretical maximum value of the distribution density of non - identifying traffic signs, β max is the theoretical maximum value of the impact of non - identifying traffic signs on the driving path, used to normalize β sign .

[0097] This step quantifies the complexity of the construction area by comprehensively weighting the distribution density and the influence coefficient. When the non - identifying traffic sign reference value NRV is relatively high, it indicates that the traffic signs in this area are relatively dense and have a strong influence, and the environmental complexity is high; conversely, a lower reference value indicates that the number of signs in this area is small and the complexity is low.

[0098] The larger the performance value of the non-sign traffic sign reference value generated after analyzing the quantity and density of the emerging informal traffic signs and temporary signs under the detection window, the higher the quantity and density of the informal traffic signs and temporary signs in the construction area. This usually means that the road environment becomes more complex and uncertain. When the number of these signs and signs increases and the distribution becomes denser, the driving environment of the vehicle becomes more dynamic and unpredictable because these temporary signs usually lack standardization and consistency, which may cause confusion or information conflicts and increase the judgment difficulty of the system. Therefore, a larger performance value of the non-sign traffic sign reference value indicates that the environment where the vehicle is located is full of uncertainty and complexity, requiring more attention and computing resources to make accurate path planning and decisions, thereby increasing the environmental complexity. If the performance value of the non-sign traffic sign reference value is small, it indicates that the number of informal signs and temporary signs in the construction area is small and the distribution is relatively scattered, and the dynamic changes in the environment are small, usually indicating that the road environment in this area is relatively simple and the environmental complexity where the vehicle is located is low.

[0099] The intelligent evaluation and classification module inputs the quantified complexity features into a pre-trained Support Vector Machine (SVM) model for intelligent evaluation, and divides the vehicle driving environment into a high-complexity environment and a low-complexity environment.

[0100] Input the quantified dynamic path blockage reference value and non-sign traffic sign reference value into a pre-trained Support Vector Machine (SVM) model. Generate an environmental complexity index through the SVM model, and conduct intelligent evaluation of the vehicle driving environment through the environmental complexity index.

[0101] The pre-trained Support Vector Machine (SVM) model refers to a model that has been able to identify, classify, or regress through machine learning algorithms for training the Support Vector Machine (SVM) under a specific dataset and task background. The Support Vector Machine is a widely used supervised learning algorithm mainly used for classification and regression problems, especially suitable for pattern recognition of small samples and high-dimensional data. In the context of applying to environmental complexity evaluation, the pre-trained SVM model will learn the relationship between these factors and the complexity of the vehicle driving environment based on historical road traffic data, construction area characteristics, non-sign traffic signs, and dynamic path blockages. Through the training process, the SVM model can map the input feature data (such as the dynamic path blockage reference value and non-sign traffic sign reference value) to a suitable output space through an optimized hyperplane or boundary, thereby automatically conducting intelligent evaluation of the complexity of the environment where the vehicle is located.

[0102] In practical applications, the pre-trained SVM model is trained on a specific labeled dataset, which contains the characteristics of various complex road environments and the corresponding environmental complexity labels (such as low-complexity environment, high-complexity environment). By training on this data, the SVM model constructs a decision boundary to distinguish different categories of environmental complexity. When new data (such as the dynamic path blockage reference value and non-identifying traffic sign reference value from the road condition perception system) is input into the model, the SVM will perform reasoning and classification based on the learned decision boundary. The core advantage of this model is that it can still maintain good performance in the context of small samples and high-dimensional data. By introducing a kernel function, the SVM can map low-dimensional input data to a high-dimensional feature space, thereby improving the classification accuracy of the model. The pre-trained SVM model usually includes a specific kernel function (such as the radial basis kernel function, linear kernel function, etc.), and adjusts the model parameters through an optimization algorithm (such as the SMO algorithm) to ensure that it can make accurate classification or regression predictions for complex input features (such as environmental blockage, non-standard signs, etc.). Therefore, the pre-trained SVM model provides an efficient and accurate way to evaluate the environmental complexity for the system, which can respond in real time in a dynamically changing traffic environment, thereby improving the intelligent level of path planning and driving decision-making.

[0103] The support vector machine model is not limited here. Any support vector machine model that can comprehensively analyze the dynamic path blockage reference value DPBF and the non-identifying traffic sign reference value NRV to generate the environmental complexity index ENVC can be used. To implement the solution of the present invention, the present invention provides a specific implementation method;

[0104] The formula for generating the environmental complexity index ENVC is as follows:

[0105] ENVC = f1·DPBF + f2·NRV

[0106] , where f1 and f2 are the preset proportionality coefficients of the dynamic path blockage reference value DPBF and the non-standard traffic sign reference value NRV respectively, and both f1 and f2 are greater than 0.

[0107] From the environmental complexity index, it can be seen that the larger the value of the dynamic path blockage reference value generated by analyzing the frequency and severity of the blocked path under the detection window, and the larger the value of the non-identifying traffic sign reference value generated by analyzing the number and density of the informal traffic signs and temporary signs that appear under the detection window, the larger the value of the environmental complexity index generated by the pre-trained support vector machine model for intelligent evaluation of the vehicle driving environment, indicating that the probability of the vehicle being in a highly complex environment is greater. On the contrary, it indicates that the probability of the vehicle being in a highly complex environment is smaller.

[0108] The preset proportionality coefficients here refer to the two coefficients f1 and f2 in the formula, and their role is to measure the comprehensive influence weights of the dynamic path blockage reference value DPBF and the non-identifying traffic sign reference value NRV on the environmental complexity index ENVC. These two coefficients are fixed values set in advance according to the requirements of the actual scenario and the results of data analysis, and are used to reflect the importance of each variable to the environmental complexity. For example, in some cases, the dynamic path blockage DPBF may have a greater impact on the environmental complexity (such as frequent roadblocks or serious construction), and at this time, the value of f1 will be relatively large; in other cases, the density of non-identifying traffic signs NRV has a greater interference on driving decisions (such as a large number of temporary signs or chaotic guiding information), and at this time, the value of f2 will be relatively large. By adjusting the values of f1 and f2, the weight requirements in different scenarios can be flexibly adapted, so as to accurately generate the ENVC reflecting the actual environmental complexity. In addition, these coefficients are usually optimized through statistical analysis of historical data or the model training process to ensure their relevance and accuracy to the actual complexity.

[0109] When the environmental complexity index generated by intelligently evaluating the vehicle driving environment through a pre-trained support vector machine model is compared and analyzed with the pre-set environmental complexity index reference threshold, the vehicle driving environment is divided, and the division steps are as follows:

[0110] If the environmental complexity index is greater than the pre-set environmental complexity index reference threshold, the vehicle driving environment is divided into a high-complexity environment;

[0111] If the environmental complexity index is less than or equal to the pre-set environmental complexity index reference threshold, the vehicle driving environment is divided into a low-complexity environment.

[0112] A high-complexity environment refers to an environment in which the vehicle is interfered by various dynamic and unpredictable factors during driving, resulting in increased driving difficulty and complicated path planning. A low-complexity environment refers to an environment in which there are fewer external interferences during vehicle driving, and the road conditions and traffic flow are relatively stable and predictable.

[0113] The stability path planning module maintains the preset data update frequency in a low-complexity environment to ensure the stability and efficiency of path planning;

[0114] The purpose of maintaining a preset data update frequency in a low-complexity environment is to ensure the efficiency and reliability of path planning and driving decisions through stable and regular data collection, while avoiding excessive consumption of system resources. In a low-complexity environment, road conditions and traffic flow are relatively stable, and external interference is less, so there is no need to update data frequently. Maintaining the preset update frequency can reduce unnecessary computational burdens on the basis of ensuring data timeliness, ensuring the stability of system operation and the continuity of path planning. At the same time, this strategy can optimize the allocation of the system's computational resources, leaving more resources for dynamic adjustment requirements in high-complexity environments, thereby improving the overall transportation efficiency and system response capabilities. By maintaining the update frequency in a low-complexity environment, cold chain logistics vehicles can complete transportation tasks in a more efficient and stable manner while ensuring the accuracy of path planning.

[0115] The dynamic path planning and real-time adjustment module dynamically increases the data update frequency in a high-complexity environment to promptly respond to rapid changes in road conditions and recalculate and optimize the vehicle's driving route.

[0116] In a high-complexity environment, the specific steps for dynamically increasing the data update frequency to promptly respond to rapid changes in road conditions and recalculate and optimize the vehicle's driving route are as follows:

[0117] When the environmental complexity index ENVC is greater than the environmental complexity index reference threshold, it indicates that the environmental complexity of the vehicle is beyond the normal range, and it is necessary to dynamically adjust the data update frequency to promptly respond to rapidly changing road conditions. At this time, based on the environmental complexity index ENVC and the preset data update frequency, calculate the new data update frequency. The calculation formula is as follows:

[0118]

[0119] , where f0 is the preset data update frequency, ENVC is the environmental complexity index, ENVC ref is the environmental complexity index reference threshold, and α is a regulation factor used to control the sensitivity of the update frequency. When the environmental complexity increases significantly, the frequency can increase exponentially. f new is the new data update frequency;

[0120] When the environmental complexity index ENVC is greater than the reference threshold, the update frequency will increase exponentially to ensure the timeliness of data collection. This dynamic adjustment process can ensure that the system can quickly respond to emergencies such as construction areas, traffic accidents, and road closures.

[0121] After adjusting the update frequency, the road condition perception system starts to collect the latest road traffic images and environmental data at the new frequency. At this time, sensors (such as cameras, radars, lidars, etc.) update data according to the adjusted frequency and transmit it in real time to the central processing system for analysis. The environmental data obtained at this time includes dynamic factors such as traffic flow, road signs, obstacles, construction information, lane width changes, and road surface damage. The latest data obtained is represented by the following formula:

[0122]

[0123] , where D current (t) is the complete environmental data set at time t, containing all sensor data, and S p (t) is the raw data provided by the p-th sensor (such as a camera, radar, etc.) at time t, and W p is the weighting coefficient of the p-th sensor data, which is set according to its importance for path planning and driving decisions. v is the total number of sensors;

[0124] This data set will provide basic support for the path planning algorithm to ensure that the system can obtain key data reflecting the current environmental changes in real time and avoid decision-making mistakes caused by lagging information updates.

[0125] After obtaining the latest environmental data, recalculate and optimize the vehicle driving route through the path planning algorithm. Considering factors such as smoothness, complexity, and safety, select the optimal path to cope with the dynamically changing road conditions.

[0126] In a high-complexity environment, the step of dynamically increasing the data update frequency plays a crucial role, mainly reflected in enhancing the system's response ability to rapidly changing environments and timely adjusting the driving path. In high-complexity environments such as construction areas, accident-prone areas, or road renovation areas, the road conditions change constantly. The traditional fixed-frequency data collection method may not be able to reflect these sudden changes in time. For example, the sudden entry of construction vehicles, temporarily closed sections, sudden traffic accidents, or temporary traffic control may all lead to drastic fluctuations in traffic flow and road conditions. At this time, if the system still updates data according to the preset fixed frequency, it may cause the system to fail to capture these key real-time changes, resulting in lagging path planning and increasing the driving risk of the vehicle.

[0127] Therefore, the core function of dynamically increasing the data update frequency is to ensure that the system can track the rapid changes in the road environment in real time, improve the sensitivity of environmental perception, and thus support timely path optimization and adjustment. When the environmental complexity increases, by shortening the data update interval, the system can obtain and process environmental information from sensors such as cameras, radars, and lidar more frequently, and identify information such as possible road obstacles, sudden traffic conditions, or road closures in real time. This enables the vehicle to quickly adjust its driving strategy, avoid congested sections or dangerous areas, thereby reducing travel time, ensuring driving safety, and avoiding delays caused by traffic jams or unforeseen accidents.

[0128] In addition, frequent data updates also provide sufficient data support for real-time path recalculation and dynamic traffic condition assessment, enabling the system to react quickly when new changes are detected and re-plan the optimal driving route. This dynamic adjustment not only ensures the efficiency of vehicle driving but also improves the timeliness and reliability of cold chain logistics in an uncertain and complex environment, avoiding delays or quality problems of goods. Therefore, dynamically adjusting the data update frequency is a key measure to cope with road condition changes in a high-complexity environment and ensure transportation efficiency and safety.

[0129] The present invention can effectively solve the problem of reaction lag caused by the fixed data collection frequency in traditional road condition perception systems by adjusting the data update frequency in real time. In this solution, the system first captures traffic images and data information of the surrounding environment through sensors, and after preprocessing and key feature extraction, accurately identifies the complexity of the environment where the vehicle is located. In a high-complexity environment, such as a construction area or a road renovation area, the system will dynamically increase the data update frequency, so as to respond in a timely manner to the rapid changes in road conditions, ensure that emergencies (such as blockages, obstacles, or temporary closures, etc.) can be captured in time, and quickly adjust the driving route to avoid delays and unnecessary detours. On the contrary, in a low-complexity environment, maintaining regular data updates can not only ensure the stable operation of the system but also reduce resource consumption. This flexible data update strategy not only improves the efficiency and accuracy of path planning but also effectively reduces the loss of timeliness, ensures the punctuality of cold chain logistics transportation and the quality of goods, and improves the customer experience. At the same time, the intelligent evaluation system enhances the ability to quickly respond to environmental changes, improves the ability to cope with complex traffic situations and uncertain factors, and optimizes the overall efficiency of the logistics process.

[0130] The above formulas are all dimensionless and take their numerical calculations. The formula is obtained by collecting a large amount of data for software simulation to get a formula closest to the real situation. The preset parameters in the formula are set by those skilled in the art according to the actual situation.

[0131] Only some exemplary embodiments of the present invention have been described by way of illustration. Undoubtedly, for those of ordinary skill in the art, without departing from the spirit and scope of the present invention, the described embodiments can be modified in various different ways. Therefore, the above drawings and description are illustrative in nature and should not be construed as limiting the scope of protection of the claims of the present invention.

[0132] It should be noted that in this document, if there are relational terms such as first and second, etc., they are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the term "comprising", "including" or any other variant thereof is intended to cover non-exclusive inclusion, such that a process, method, article or device comprising a series of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article or device. Without further limitation, an element defined by the statement "comprising one..." does not exclude the presence of another identical element in the process, method, article or device comprising the element.

[0133] It should be understood that in various embodiments of the present application, the magnitude of the sequence numbers of the above processes does not mean the order of execution is prior or posterior. The order of execution of each process should be determined by its function and internal logic, and should not constitute any limitation to the implementation process of the embodiments of the present application.

[0134] Those of ordinary skill in the art can realize that the units and algorithm steps of each example described in combination with the embodiments disclosed herein can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are executed in a hardware or software manner depends on the specific application and design constraints of the technical solution. Professional technicians can use different methods for each specific application to implement the described functions, but such implementation should not be considered to exceed the scope of the present application.

[0135] Those skilled in the art can clearly understand that for the sake of convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the foregoing method embodiments and will not be elaborated herein.

[0136] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0137] In addition, in each embodiment of the present application, each functional unit may be integrated in a processing unit, may exist separately as individual physical units, or two or more units may be integrated in one unit.

[0138] As described above, the foregoing are only specific embodiments of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed in the present application, and all such changes or substitutions should be covered by the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.

[0139] Only some exemplary embodiments of the present invention have been described above by way of illustration. Undoubtedly, for those of ordinary skill in the art, the described embodiments can be modified in various different ways without departing from the spirit and scope of the present invention. Therefore, the above drawings and description are illustrative in nature and should not be construed as limiting the protection scope of the claims of the present invention.

Claims

1. A cold chain logistics vehicle route planning system based on data analysis, characterized in that, It includes a data acquisition and sensor scanning module, a data preprocessing and feature extraction module, an environmental complexity quantification and analysis module, an intelligent evaluation and classification module, a stability path planning module, and a dynamic path planning and real-time adjustment module: Data acquisition and sensor scanning module: First, the road condition perception system regularly scans the surrounding environment through the sensors on the vehicle at a preset regular data update frequency to obtain real-time road traffic images and environmental data information, providing stable data support for subsequent path planning and driving decisions; Data preprocessing and feature extraction module: After obtaining the original road traffic images and environmental data information, preprocess the acquired data and extract the key features reflecting that the current vehicle is in a high-complexity environment from the preprocessed data; Environmental complexity quantification and analysis module: In the detection window, deeply analyze the extracted key features and quantify the complexity of the vehicle driving environment through the analyzed features; Intelligent evaluation and classification module: Input the quantified complexity features into a pre-trained support vector machine model for intelligent evaluation, and classify the vehicle driving environment into a high-complexity environment and a low-complexity environment; Stability path planning module: In a low-complexity environment, maintain the preset data update frequency to ensure the stability and efficiency of path planning; Dynamic path planning and real-time adjustment module: In a high-complexity environment, dynamically increase the data update frequency, promptly respond to the rapid changes in road conditions, and recalculate and optimize the vehicle's driving route.

2. The cold chain logistics vehicle route planning system based on data analysis according to claim 1, characterized in that, Extract the key features reflecting that the current vehicle is in a high-complexity environment from the preprocessed data. The extracted features include the frequency and severity of path blockage, as well as the number and density of informal traffic signs and temporary signs that appear. In the detection window, analyze the frequency and severity of path blockage and the number and density of informal traffic signs and temporary signs that appear to generate a dynamic path blockage reference value and a non-identifying traffic sign reference value, and quantify the complexity of the vehicle driving environment through the dynamic path blockage reference value and the non-identifying traffic sign reference value.

3. The cold chain logistics vehicle route planning system based on data analysis according to claim 2, characterized in that, The specific steps for analyzing the frequency and severity of path blockage in the detection window to generate a dynamic path blockage reference value are as follows: First, define a path blockage event. To quantify the frequency, set the blockage event occurrence rate, which represents the number of blockage events per unit time. The formula for the blockage event occurrence rate is as follows: Wherein, B freq is the path blocking frequency, n is the total number of blocking events occurring within the detection window, D i is the duration of the i-th path blocking event, P i is the blocking type of the i-th path blocking event, is an indicator function used to mark whether the path blocking event i is a valid path blockage; To further improve the quantification accuracy of path blockage, identify the blockage severity and calculate the blockage impact factor. The calculation formula is as follows: Wherein, I block is the blocking influence factor, A j is the area of the j-th blocking area, V j is the traffic flow deceleration factor of the j-th blocking area, P j is the path blocking state of the j-th blocking area, is the indicator function used to mark whether there is an effective blocking event in this area; Comprehensive path blockage frequency B freq and blockage influence factor I block , to generate a dynamic path blockage reference value for representing the complexity of the vehicle's environment. The generation formula is as follows: Wherein, DPBF is the dynamic path block reference value, I block，k is the path block frequency in the k-th time period, I block，k is the path block impact factor in the k-th time period, p is the total number of time periods within the monitoring window, θ is the block impact sensitivity adjustment parameter, is the exponential weighting of the block severity.

4. A cold chain logistics vehicle route planning system based on data analysis according to claim 2, characterized in that, The specific steps for analyzing the number and density of informal traffic signs and temporary signs that appear in the detection window to generate a non-identifying traffic sign reference value are as follows: First, collect the data on the number and density of all informal traffic signs and temporary signs. Since the distribution density of informal traffic signs and temporary signs is different in different regions, a local density function is needed to analyze the density. The formula for the local density function is as follows: Wherein, ρ(x, y) is the local density function, representing the density of informal traffic signs in the area around the position (x, y), N(x, y) is the number of informal traffic signs around the position (x, y), and A(x, y) is the area of the area at the position (x, y); Combined with the local density function ρ(x, y) and the total number of signs, calculate the distribution factor of non-identifying traffic signs. The calculation expression is as follows: where DI is the distribution factor of non - identifying traffic signs, ρ(x q , y q ) is the local density of the q - th sign, N is the total number of informal traffic signs, A is the total area of the region, d(x q ) is the distance from the q - th sign to the center of the region, and d max is the maximum distance of the region, representing the maximum distance between any sign in the region and the center of the region; After obtaining the distribution factor DI of non-identifying traffic signs, further quantify the potential impact of signs on the vehicle driving path through the non-identifying traffic sign influence coefficient. The calculation formula of the non-identifying traffic sign influence coefficient is as follows: where β sign is the influence coefficient of non - iconic traffic signs, and w q is the weight coefficient of the q - th sign; Integrate the distribution factor DI of non-identifying traffic signs and the influence coefficient β of non-identifying traffic signs sign Generate the reference value of non-identifying traffic signs, and the generation formula is as follows: wherein, R sign is the reference value of non - iconic traffic signs, D max is the theoretical maximum value of the distribution density of non - iconic traffic signs, and β max is the theoretical maximum value of the influence of non - iconic traffic signs on the driving path.

5. A cold chain logistics vehicle route planning system based on data analysis according to claim 2, characterized in that, Input the quantified dynamic path blockage reference value and non-identifying traffic sign reference value into a pre-trained support vector machine model, generate an environmental complexity index through the support vector machine model, and intelligently evaluate the vehicle driving environment through the environmental complexity index.

6. The cold chain logistics vehicle route planning system based on data analysis according to claim 5, characterized in that Compare and analyze the environmental complexity index generated when the vehicle driving environment is intelligently evaluated by the pre-trained support vector machine model with the pre-set environmental complexity index reference threshold, and divide the vehicle driving environment. The division steps are as follows: If the environmental complexity index is greater than the pre-set environmental complexity index reference threshold, the vehicle driving environment is divided into a high-complexity environment; If the environmental complexity index is less than or equal to the pre-set environmental complexity index reference threshold, the vehicle driving environment is divided into a low-complexity environment.

7. An cold chain logistics vehicle route planning system based on data analysis according to claim 6, characterized in that, In a high-complexity environment, dynamically increase the data update frequency, respond in a timely manner to the rapid changes in road conditions, and recalculate and optimize the vehicle driving route. The specific steps are as follows: When the environmental complexity index ENVC is greater than the environmental complexity index reference threshold, it indicates that the environmental complexity of the vehicle's location has exceeded the normal range, and the data update frequency needs to be dynamically adjusted to respond in a timely manner to the rapidly changing road conditions. At this time, based on the environmental complexity index ENVC and the pre-set data update frequency, calculate the new data update frequency. The calculation formula is as follows: where f0 is the preset data update frequency, ENVC is the environmental complexity index, and ENVC ref is the reference threshold of the environmental complexity index, α is the adjustment factor used to control the sensitivity of the update frequency, and f new is the new data update frequency; After adjusting the update frequency, the road condition perception system starts to collect the latest road traffic images and environmental data at the new frequency. At this time, the sensor updates the data according to the adjusted frequency and transmits it in real time to the central processing system for analysis. The latest data obtained is expressed by the following formula: where D current (t) is the complete environmental data set at time t, including all sensor data, and S p (t) is the raw data provided by the p-th sensor at time t, and W p is the weighting coefficient of the p-th sensor data, and v is the total number of sensors; After obtaining the latest environmental data, recalculate and optimize the vehicle driving route through the path planning algorithm, and select the optimal path to cope with the dynamically changing road conditions.

Images