Carbon emission management system for traffic transportation industry based on Internet of Things

Abstract

The invention relates to the technical field of carbon emission management, and particularly discloses a traffic transportation industry carbon emission management system based on the Internet of Things, and the system comprises an acquisition module which divides a road into road segments, and continuously collects the illegal behavior data of non-motor vehicles on the road segments through an Internet of Things sensing device; the aggregation module is used for dynamically aggregating the violation behavior data of the road section and generating a comprehensive violation strength index of the road section in a continuous time period; the dividing module is used for dividing the road sections into different risk levels based on the comprehensive violation intensity index and in combination with the phase data of the traffic signal lamp at the current moment; and the recommendation module is used for recommending a path with the lowest total risk value according to a motor vehicle navigation request obtained in real time, dynamically receiving a non-motor vehicle real-time movement track uploaded by the Internet of Things sensing equipment in a navigation process, and triggering re-evaluation of a path risk level and real-time adjustment of a navigation path. The carbon emission in the highway transportation process is reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of carbon emission management, and in particular to a carbon emission management system for the transportation industry based on the Internet of Things. Background Art

[0002] Transportation refers to the business activity of using transportation tools to deliver goods or passengers to the destination and transfer their spatial location. At present, according to different transportation tools, transportation can be divided into five types: road transportation, railway transportation, water transportation, air transportation, and pipeline transportation. Among them, road transportation is the most common mode of transportation, with the characteristics of strong adaptability and direct transportation.

[0003] Vehicles emit carbon during driving. Frequent starts and stops, sudden acceleration and deceleration will significantly increase carbon emissions per unit mileage compared to driving at a constant speed. On urban roads, some non-motor vehicles such as electric bicycles and tricycles do not obey traffic lights, drive in the opposite direction, change lanes at will, or cross traffic, forcing the motor vehicles behind to brake suddenly, stop briefly, and then start again. The amount of fuel injected increases when the engine is increased from idle to load, and the combustion temperature fluctuates, causing the emission peaks of greenhouse gases such as carbon dioxide and nitrous oxide to overlap; at the same time, the vehicle inertia is repeatedly interrupted, resulting in an increase in overall fuel consumption. Summary of the invention

[0004] The purpose of the present invention is to provide a carbon emission management system for the transportation industry based on the Internet of Things to solve the above technical problems.

[0005] The purpose of the present invention can be achieved through the following technical solutions: A carbon emission management system for the transportation industry based on the Internet of Things, comprising: Collection module: Divide the road into road segments, and continuously collect the violation data of non-motor vehicles on the road segments through IoT sensing devices. The violation data includes the number of red light running, the number of wrong-way driving, and the number of lane change and weaving; Aggregation module: dynamically aggregates the violation data of a road segment to generate a comprehensive violation intensity index for the road segment in a continuous time period; Classification module: based on the comprehensive violation intensity index and combined with the traffic light phase data at the current moment, the road segment is divided into different risk levels; Recommendation module: Based on the real-time motor vehicle navigation requests, it recommends the path with the lowest total risk value, and dynamically receives the real-time motion trajectory of non-motor vehicles uploaded by the IoT sensing device during the navigation process, triggering the re-evaluation of the path risk level and real-time adjustment of the navigation path.

[0006] As a further solution of the present invention: collecting illegal behavior data of non-motor vehicles on a road segment includes: Install multi-view monitoring equipment at a preset location to capture the displacement distance and duration of non-motor vehicles crossing the stop line during a red light through image recognition technology. When the displacement distance exceeds the preset displacement distance threshold and continues until the green light comes on without returning, it will be recorded as running a red light; Millimeter-wave radars are installed on the middle isolation barriers of the road to monitor the relative speed difference and lateral displacement changes between non-motor vehicles and motor vehicles in adjacent lanes in a single-direction traffic flow. When a non-motor vehicle is detected to continuously cross more than two lane lines within a preset length range, it is recorded as a lane change; Through the pressure sensor array buried at the edge of the non-motorized vehicle lane, the continuous pressure signal sequence generated when the non-motorized vehicle travels along a path opposite to the prescribed driving direction is identified, and the reverse driving is determined based on the time interval and spatial distribution characteristics of the pressure signal.

[0007] As a further solution of the present invention: generating a comprehensive violation intensity index of a road segment in a continuous time period includes: The violation data of a single road section is divided into multiple time windows according to the preset duration, and the number of red light running, lane changing and wrong-way driving in a single time window is counted; Different weight coefficients are assigned to the three types of violation behavior data in the same time window for weighted summation to obtain the initial intensity value, among which the weight of red light running is the highest, the weight of wrong-way driving is the second, and the weight of lane changing is the lowest. A moving average is calculated based on the initial intensity values of the same time window of the road segment within the past preset number of days, and the ratio of the initial intensity value of the current time window to the moving average is used as the comprehensive violation intensity index.

[0008] As a further solution of the present invention: dividing the road segments into different risk levels includes: The risk levels include high risk, medium risk and low risk; When the number of time windows in which the comprehensive violation intensity index continuously exceeds the preset threshold reaches a preset first number, and the traffic light is at the end of the red light or the beginning of the green light at the current moment, the corresponding risk level is high risk; When the comprehensive violation intensity index exceeds the preset threshold within a single time window but does not meet the continuous condition and / or the traffic light is in the yellow light switching stage, the corresponding risk level is medium risk; When the comprehensive violation intensity index does not exceed the preset threshold and the signal light is in the middle of the green light, the corresponding risk level is low risk.

[0009] As a further solution of the present invention: generating a comprehensive violation intensity index of a road segment in a continuous time period includes: Establish a communication connection with the traffic signal controller through the roadside IoT gateway to read the current phase status and remaining duration of the traffic light in real time; When the signal light is in the red light state, synchronously obtain the cumulative duration from the moment when the red light is turned on to the current moment and the preset total red light duration; When the signal light is in the green light state, synchronously obtain the cumulative duration from the moment when the green light is turned on to the current moment and the preset remaining green light duration.

[0010] As a further solution of the present invention: The path with the lowest recommended total risk value includes: When receiving a motor vehicle navigation request, extract all feasible path sets from the departure place to the destination; Eliminate the paths that contain high-risk road segments and have no alternative detour plans from the path set; For the remaining paths, calculate the total path risk value according to the risk levels of each road segment, where the high-risk road segments are assigned the maximum risk coefficient, the medium-risk road segments are assigned the intermediate risk coefficient, and the low-risk road segments are assigned the minimum risk coefficient; After arranging the total path risk values in ascending order, recommend the path with the lowest total risk value, and mark the risk level distribution on the path in different colors on the electronic map interface.

[0011] As a further solution of the present invention: Triggering the re-evaluation of the path risk level includes: It is detected by the Internet of Things sensing device that in a certain road segment of the currently recommended path, the density of the real-time movement trajectories of non-motor vehicles exceeds the historical same-period level; Or it is detected that there is a sudden traffic control at the upstream intersection of the road segment, resulting in abnormal convergence of non-motor vehicle flows; Or it is detected that there are group non-motor vehicles continuously performing lane-changing and cutting-in behaviors within the road segment; When any triggering condition is met, re-call the violation behavior data within the recently set time period for aggregation calculation, update the comprehensive violation intensity index and risk level of the road segment, and generate a new path recommendation instruction according to the updated result.

[0012] As a further solution of the present invention: The real-time adjustment of the navigation path includes: When the vehicle is driving along the recommended path, the Internet of Things sensing device detects that the risk level of the front road segment increases, and triggers the path dynamic switching mechanism; Based on the current position of the vehicle, recalculate the alternative path set, and the alternative path set needs to meet the following conditions: exclude all newly elevated high-risk road segments, and the total length of the detour path does not exceed a set ratio of the original path length; Select the new path with the lowest total risk value from the alternative path set, and push a path change request to the driver through the in-vehicle terminal; If the driver confirms to accept the change, the navigation guidance trajectory is updated, and a steering instruction sequence is sent to the vehicle control system through the Internet of Vehicles. The steering instruction sequence includes the lane change prompts and steering angle suggestions required before each intersection; If the driver does not respond to the request, a warning signal will be broadcast to the vehicle behind through the Internet of Vehicles communication, triggering the vehicle behind to automatically maintain a safe distance. Based on the predicted motion trajectory of the non-motor vehicle in front, speed control instructions will be dynamically generated to guide the vehicle through high-risk road sections with a gentle acceleration and deceleration curve.

[0013] The beneficial effects of the present invention are as follows: The present invention collects data on violations such as red light running, driving in the wrong direction, and lane change in real time by deploying IoT sensing devices on road sections, and dynamically calculates the comprehensive violation intensity index in combination with the phase of traffic lights, and divides the road sections into high, medium, and low risk levels; the navigation link eliminates or weakens high-risk road sections during path planning, and automatically re-evaluates risks and adjusts the driving route in a rolling manner according to the real-time violation situation during driving. As a result, motor vehicles can avoid violation-intensive sections in advance, maintain a more stable speed curve, and significantly reduce the additional power output and fuel consumption caused by emergency braking, frequent start-stop, and rapid acceleration-deceleration, thereby reducing the instantaneous greenhouse gas emission peak and reducing overall carbon emissions, while improving road traffic efficiency and driving safety. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] The present invention will be further described below in conjunction with the accompanying drawings.

[0015] Figure 1 It is a flow chart of a carbon emission management system for the transportation industry based on the Internet of Things according to the present invention. DETAILED DESCRIPTION

[0016] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0017] See also Figure 1 As shown, the present invention is a carbon emission management system for the transportation industry based on the Internet of Things, comprising: Collection module: Divide the road into road segments, and continuously collect the violation data of non-motor vehicles on the road segments through IoT sensing devices. The violation data includes the number of red light running, the number of wrong-way driving, and the number of lane change and weaving; In a preferred embodiment of the present invention, collecting illegal behavior data of non-motor vehicles on a road segment includes: Install two to three multi-view high-definition cameras about 3m to 5m above the stop line at the intersection. Through perspective correction, set the stop line and the area 0.5m before and after it as the region of interest. The system uses a deep learning object detection network to perform instance segmentation and trajectory tracking on non-motor vehicles such as electric bicycles and tricycles entering this region, and calculates the shortest distance from its centroid to the stop line in real time; when the cumulative centroid displacement exceeds a preset displacement threshold (the displacement threshold can be 0.30m to 0.50m) and the duration of this state from the red light on to before the green light on is greater than the preset duration threshold (the duration threshold can be 2s to 3s) and the vehicle does not retreat behind the stop line, write a red light running record into the database; Install a narrow-beam millimeter-wave radar on a fixed bracket about 1.2m above the ground of the road central isolation fence. The working frequency band of the radar can be 77GHz. The detection sector covers two to three lanes in the same direction as motor vehicles and outputs the lateral displacement Δy and relative speed Δv of non-motor vehicle targets in real time. The background calculates the instantaneous lane number of the vehicle according to the lane line equation calibrated in the electronic map; when the target crosses two or more lane lines during continuous lateral movement and the total crossing length exceeds the preset length threshold (the length threshold can be 8m to 12m), register it as a lane-changing and cutting behavior once; Bury single-point piezoelectric pressure sensors at intervals of 0.5m along the right edge of the non-motor vehicle lane to form an array. The sensor numbers increase in the specified driving direction of the vehicle. The system compares the pressure trigger sequence numbers in chronological order. When it is detected that at least 5 consecutive numbers show a decreasing trend and the adjacent trigger intervals are less than the preset time threshold (the time threshold can be 0.4s to 0.6s), after confirming that it belongs to the same target in combination with the vehicle size model, identify it as a reverse behavior once and record it; It should be noted that by using multi-view cameras, millimeter-wave radars and pressure sensors to perform complementary detection of non-motor vehicle violation behaviors at different spatial scales and dimensions, it can maintain a high recognition accuracy in complex scenarios such as light changes, rain and fog occlusion, or traffic congestion, and avoid missed detection or false detection by a single sensing means; record the violation behaviors in the database in real time according to clear and unified event rules, so that the back-end aggregation module can generate a comprehensive violation intensity index in time and feedback it to the navigation decision-making, thereby helping the system quickly identify high-risk road sections, guide motor vehicles to detour or decelerate in advance, reduce the occurrence probability of sudden braking, idling and secondary acceleration caused by non-motor vehicle sudden behaviors, and ultimately contribute to the smoothness of vehicle driving conditions and the refined control of transportation carbon emissions; Aggregation module: Dynamically aggregate the violation behavior data of the road section to generate a comprehensive violation intensity index of the road section in a continuous time period; Another preferred embodiment of the present invention, generating a comprehensive violation intensity index of the road section in a continuous time period includes: An independent buffer zone is established for each road section. The red light running, wrong-way driving, and lane changing events reported by the sensing terminal in real time are written according to the time of occurrence and automatically attributed to a time window with a preset length. The time window can be 5min to 10min and supports sliding updates of 0.5 times the window length. After the number of times N1, N2, and N3 of the three types of events in a single window are counted respectively, the weight coefficients preset in the configuration file are called for weighting, where the red light running weight w1 can be 0.5-0.6, the wrong-way weight w2 can be 0.25-0.35, and the lane changing weight w 3 can be 0.15-0.25, satisfying w1+w2+w3=1, and the initial intensity value is obtained by the formula S0=w1N1+w2N2+w3N3; then the historical initial intensity sequence of the road section in the same time window in the past preset number of days is retrieved from the database, and the preset number of days can be 7d-14d. The benchmark mean S1 is obtained by the simple moving average algorithm, and the ratio R of the current window intensity to the benchmark mean is calculated, and R is written into the time series database as the comprehensive violation intensity index, while S0 and the original event count are retained for visualization and model backtracking; It is understandable that fixed time windows can capture the instantaneous aggregation of violations in a fine-grained manner and avoid information dilution caused by long-term statistics. The weight coefficient highlights high-risk factors such as running red lights, which can increase the sensitivity of the index to key factors that cause sudden braking and fuel consumption fluctuations of motor vehicles. The dynamic benchmark based on the historical same-window moving average can automatically offset the background changes caused by weather, holidays and traffic fluctuations during the period, making the index abnormality assessment more objective. Finally, the scales of different road sections are unified in the form of ratios to facilitate horizontal comparison and threshold judgment, thereby providing real-time, stable and explainable input for risk level classification and route recommendation, helping the system to promptly discover potential congestion and high emission factors and guide vehicles to make energy-efficient and friendly driving decisions. Classification module: based on the comprehensive violation intensity index and combined with the traffic light phase data at the current moment, the road segment is divided into different risk levels; In another preferred embodiment of the present invention, dividing the road segments into different risk levels comprises: The risk levels include high risk, medium risk and low risk; Maintain a sequence of real-time updated comprehensive violation intensity indices for each road segment in the edge computing server. First, the system reads the thresholds R0 and N0 from the configuration file. R0 can be in the range of 1.3 to 1.5, and N0 can be in the range of 3 to 4. The algorithm traverses the last N0 windows in steps of 1 time window. When it is detected that the indices of these N0 windows are all greater than or equal to R0, it enters the lamp color discrimination process: If the roadside signal controller returns that the current lamp color is red and the remaining red light duration Tr ≤ 20% of the total red light duration, it is defined as the late stage of the red light; if the lamp color is green and the lit duration Tg ≤ 10% of the total green light duration, it is defined as the early stage of the green light. When the index is continuously above the threshold and the lamp color is in the above-mentioned late stage of the red light or early stage of the green light, write the risk level of the road segment as high risk; if the index ≥ R0 in any single window but the number of consecutive windows is less than N0, or the lamp color is yellow and the lit duration Ty ≤ 50% of the total yellow light duration, write it as medium risk; if the index of the current window is less than R0, the lamp color is green, and the lit duration of the green light is between 30% and 70% of the total green light duration, that is, it is defined as the middle stage of the green light, then write it as low risk. Subsequently, the system encapsulates the risk level together with the timestamp into a JSON message and pushes it to the route recommendation engine; It should be noted that the continuous exceeding of the threshold of the comprehensive violation intensity index represents that there are high-density high-risk violation behaviors continuously existing on the road segment for a period of time. It is necessary to superimpose with the signal phase to accurately evaluate the immediate impact on motor vehicle traffic. The late stage of the red light and the early stage of the green light are the stages where the behaviors of people and vehicles change most violently. Non-motor vehicles often make sudden crossroads in order to rush through or accelerate through the intersection, and motor vehicles are in the state of about to start or just starting. The superposition of the two is most likely to trigger sudden braking and sudden acceleration. Therefore, it is defined as high risk; if the index only exceeds the threshold in a single window or the lamp color is in the yellow light switching state (that is, the lamp color is yellow and the lit duration Ty ≤ 50% of the total yellow light duration), at this time, the violation behavior or signal change has not formed a continuous impact, and the risk is obvious but can be controlled by decelerating and concentrating attention, so it is classified as medium risk; a low index and the middle stage of the green light mean that the non-motor vehicle flow and motor vehicle traffic have tended to be stable and it is not easy to trigger sudden braking, so it is classified as low risk. This dual discrimination based on intensity persistence and lamp color stage makes the risk classification both sensitive and stable, and can provide targeted decision-making basis for navigation dynamic detouring and vehicle adaptive speed control, so as to help the system remind drivers or autonomous driving systems to prevent non-motor vehicle interference in the most critical time window, reduce sudden stops and starts, and support the realization of the goal of fine management of traffic carbon emissions; It should be noted that for ordinary road segments without traffic signal control, the risk level is directly divided according to the absolute value range of the comprehensive violation intensity index. The specific division range can be set according to experience and is not limited here; In a preferred case of this embodiment, dividing the road segment into different risk levels further includes: The roadside Internet of Things gateway establishes a two-way long connection with the traffic signal controller through Gigabit Ethernet or RS-485 bus, and uses the NTCIP or JT / T794 protocol to send status query messages at a polling period of 0.5s to 1s and parse the phase number, lamp color identification, and countdown information in the returned frame. When the currently parsed lamp color is red, the built-in phase table of the controller is immediately read, the timestamp T0 when this red light starts to light is extracted, and the cumulative duration Δt1 from T0 to the query time T1 is calculated using the local high-precision clock. At the same time, the preset total red light duration L_r of this phase in the timing plan is retrieved. If the parsed result is green, the green light start timestamp T2 is similarly recorded and the cumulative duration Δt2 is calculated, and compared with the remaining green light duration L_g_r in the timing plan. If a difference exceeding the preset deviation threshold (the deviation threshold can be 0.5s to 1s) is found between the countdown and the local calculation, a secondary synchronization with the controller is triggered and the clock offset is updated. All obtained phase states, cumulative durations, and remaining durations are encapsulated into a JSON object and published to the edge computing server via MQTT for subsequent calls; It should be noted that using real-time high-frequency communication to ensure that the system grasps the most accurate signal lamp phase and remaining time information can quickly judge risk scenarios at the key nodes where non-motor vehicle violations are connected to the signal phase. Using local clock difference correction and countdown comparison can improve data consistency and avoid phase misjudgment caused by clock drift of the controller; providing a unified scale through the preset total red light duration and remaining green light duration thresholds facilitates the normalization of indicators and rule triggering at different intersections, so that the comprehensive violation intensity index is more accurately linked to the signal phase, providing a reliable timing basis for the subsequent risk level division and dynamic adjustment of navigation, ultimately helping vehicles anticipate upcoming changing traffic conditions in advance, reducing sudden stops and starts, and supporting the realization of the fine control goal of carbon emissions; Recommendation module: According to the real-time obtained motor vehicle navigation request, recommend the path with the lowest total risk value, and dynamically receive the real-time movement trajectories of non-motor vehicles uploaded by Internet of Things sensing devices during the navigation process, triggering a re-evaluation of the path risk level and real-time adjustment of the navigation path; Another preferred embodiment of the present invention, the recommended path with the lowest total risk value includes: After the in-vehicle terminal or the cloud path planning module receives a navigation request, it first calls the map engine to search for K feasible paths from the starting point to the ending point in the "road segment - topology" model. K can be 5 to 8. Each path consists of the numbers of several road segments and their attributes such as length and speed limit. Subsequently, it checks each path one by one to see if it contains nodes that have been marked as high-risk by the risk division module and have no available alternative road segments within a lateral search radius of 500m. If any exist, the path is directly removed from the candidate set. For the remaining paths, a risk coefficient is assigned to each road segment: the high-risk coefficient can be 2.5 to 3.0, the medium-risk coefficient can be 1.5 to 2.0, and the low-risk coefficient can be 1.0 to 1.2. Then, a weighted sum is performed to obtain the total risk value of the path. The total risk values of all candidate paths are sorted in ascending order, and the path with the minimum value is taken as the recommended result. At the same time, on the HMI interface, the GIS rendering service is used to color the low-risk segments green, the multi-risk segments yellow, and the high-risk segments red, and transparent dots are superimposed at each intersection to indicate risk changes within the next 50m. It should be noted that excluding high-risk paths with no room for detour can prevent navigation from leading the vehicle into a potential bottleneck area and being unable to escape in time. Multiplying the risk coefficient by the road segment length and accumulating them can normalize different driving distances and risk levels to the same evaluation scale, reflecting both the risk intensity and taking into account the driving cost. By setting K candidate paths, the calculation amount is restricted to ensure real-time performance, and setting specific risk coefficient intervals leaves room for flexible parameter adjustment according to urban management strategies in the later stage. Finally, visually marking the risk distribution with colors can help drivers or autonomous driving systems perceive dangerous sections in advance and adjust driving strategies smoothly, thereby reducing the probability of sudden lane changes and emergency braking, and providing an operable navigation basis for reducing carbon emissions and improving traffic safety.

[0018] In a preferred case of this embodiment, triggering a re-evaluation of the path risk level includes: In the edge computing server, the number of non-motor vehicle trajectories continuously reported by the millimeter-wave radar is summarized in a five-second sliding window and written into the time series database, and the average number of trajectories in the same window at the same time period in the past fourteen days is retrieved. When the real-time number is greater than 1.5 times the average value and this condition is met for three consecutive windows, it is determined that the trajectory density of this road segment is abnormal. Through the event interface with the traffic management cloud platform, the signal machine status of the intersection within 100 meters upstream of the target road segment is subscribed to in real time. If it is found that the intersection enters a state of temporary red light always on or full red flash and lasts for more than 30 seconds, it is considered that sudden traffic control has led to traffic convergence. Spatial clustering is performed on the target trajectories captured by the millimeter-wave radar and high-definition cameras within the road segment. When at least five non-motor vehicles are detected within a 20-meter length range, with the cumulative lateral displacement exceeding 1 meter and continuously crossing two lane lines within three seconds, it is determined that a group lane-changing and cutting behavior has occurred. Once any condition is triggered, the system immediately calls the data of red-light running, reverse driving, lane-changing, and cutting-in events within the last five minutes for re-aggregation to calculate the comprehensive violation intensity index, updates the risk level of the road segment, and issues an instruction containing the new level to the route recommendation engine. The in-vehicle terminal then pushes a route adjustment prompt. It can be understood that by real-time monitoring of abnormal trajectory density, upstream flow mutation, and capturing group lateral movement behavior, the system can identify early signals of increased risk before a large accumulation of violation events. The combination of sliding window statistics and multiple thresholds ensures the sensitivity of the detection to instantaneous fluctuations while suppressing short-term noise. Limiting the five-minute data recalculation avoids the slow response caused by historical lag. Once the risk level is refreshed, it drives the navigation to quickly detour or reminds to slow down, reducing frequent hard braking, starting and stopping, and rapid acceleration from the source, providing timely and reliable decision-making basis for the goals of smooth driving and fine control of carbon emissions. Another preferred case of this embodiment, the real-time adjustment of the navigation path includes: Subscribe to the latest risk level messages of the road segment within the next 200 meters ahead at a one-second interval. When it is detected that the road segment changes from medium risk to high risk or directly from low risk to high risk, immediately trigger the path dynamic switching mechanism. The navigation engine re-searches the candidate path set from the starting point to the destination with the vehicle's current position as the starting point. When searching, it performs real-time weighting on the map data and excludes all road segments marked as high risk and difficult to downgrade within ten minutes, and requires that the total length of the newly generated detour path does not exceed 1.15 to 1.3 times the length of the original path. Call the risk weight calculation function for each candidate path that meets the requirements to obtain the total risk value of the path, select the one with the smallest total risk value, and push a path change request through the in-vehicle terminal in a dual-channel manner of voice and the central control screen. If the driver clicks to confirm or confirms by voice within five seconds, the vehicle networking immediately updates the navigation trajectory and continuously sends a sequence of steering instructions to the vehicle control system. Each instruction is issued when the vehicle is 50 meters away from the intersection, including the lane number recommended to change to, the recommended steering angle, and the early speed change prompt. If no driver response is received within five seconds, the vehicle networking broadcasts a warning message to the connected vehicles within 100 meters behind through dedicated short-range communication technology, triggering the following vehicles to automatically maintain a safety distance of no less than 30 meters. At the same time, according to the predicted results of the future five-second movement trajectory of the non-motor vehicles ahead by the sensor, a longitudinal speed control curve is generated and the acceleration change rate is limited not to exceed 0.2 gravitational acceleration to guide the vehicle to pass through the high-risk road segment in a smooth manner. It should be noted that by immediately recalculating the alternative path and restricting the detour length when the risk level suddenly rises, it is possible to quickly avoid high-risk areas on the premise of ensuring that the travel distance and time cost are acceptable. Sending clear route and control instructions to the driver in real time can reduce the delay of manual decision-making. When the driver fails to give timely feedback, automatically issuing a warning and coordinating the following vehicle to maintain a safe distance and a smooth speed curve can ensure more stable longitudinal power output of the vehicle, reduce sudden lane changes and hard braking, and generally help to maintain a uniform driving condition, reduce the additional fuel consumption and greenhouse gas emissions caused by traffic shocks, and further improve driving safety and carbon emission management effects; The above formulas are all dimensionless and take their numerical calculations. The formulas are obtained by collecting a large amount of data for software simulation to get a formula closest to the real situation. The preset parameters and threshold selection in the formulas are set by those skilled in the art according to the actual situation.

[0019] The above has described in detail an embodiment of the present invention, but the content described is only a preferred embodiment of the present invention and cannot be considered as limiting the scope of implementation of the present invention. All equivalent changes and improvements made within the scope of the application of the present invention should still fall within the scope covered by the present invention.

Claims

1. An Internet of Things-based carbon emission management system for the transportation industry, characterized in that, include: Collection module: Divide the road into road segments, and continuously collect the violation data of non-motor vehicles on the road segments through IoT sensing devices. The violation data includes the number of red light running, the number of wrong-way driving, and the number of lane change and weaving; Aggregation module: dynamically aggregates the violation data of a road segment to generate a comprehensive violation intensity index for the road segment in a continuous time period; Classification module: based on the comprehensive violation intensity index and combined with the traffic light phase data at the current moment, the road segment is divided into different risk levels; Recommendation module: Based on the real-time motor vehicle navigation requests, it recommends the path with the lowest total risk value, and dynamically receives the real-time motion trajectory of non-motor vehicles uploaded by the IoT sensing device during the navigation process, triggering the re-evaluation of the path risk level and real-time adjustment of the navigation path.

2. The carbon emission management system for the transportation industry based on the Internet of Things according to claim 1, wherein The data collected on violations of non-motor vehicles on road sections include: Install multi-view monitoring equipment at a preset location to capture the displacement distance and duration of non-motor vehicles crossing the stop line during a red light through image recognition technology. When the displacement distance exceeds the preset displacement distance threshold and continues until the green light comes on without returning, it will be recorded as running a red light; Millimeter-wave radars are installed on the middle isolation barriers of the road to monitor the relative speed difference and lateral displacement changes between non-motor vehicles and motor vehicles in adjacent lanes in a single-direction traffic flow. When a non-motor vehicle is detected to continuously cross more than two lane lines within a preset length range, it is recorded as a lane change; Through the pressure sensor array buried at the edge of the non-motorized vehicle lane, the continuous pressure signal sequence generated when the non-motorized vehicle travels along a path opposite to the prescribed driving direction is identified, and the reverse driving is determined based on the time interval and spatial distribution characteristics of the pressure signal.

3. An Internet of Things-based carbon emission management system for the transportation industry according to claim 1, characterized in that, Generating a comprehensive violation intensity index for a road segment over a continuous time period includes: The violation data of a single road section is divided into multiple time windows according to the preset duration, and the number of red light running, lane changing and wrong-way driving in a single time window is counted; Different weight coefficients are assigned to the three types of violation behavior data in the same time window for weighted summation to obtain the initial intensity value, among which the weight of red light running is the highest, the weight of wrong-way driving is the second, and the weight of lane changing is the lowest. A moving average is calculated based on the initial intensity values of the same time window of the road segment within the past preset number of days, and the ratio of the initial intensity value of the current time window to the moving average is used as the comprehensive violation intensity index.

4. The carbon emission management system for the transportation industry based on the Internet of Things according to claim 1, characterized in that, The classification of road segments into different risk levels includes: The risk levels include high risk, medium risk and low risk; When the number of time windows in which the comprehensive violation intensity index continuously exceeds the preset threshold reaches a preset first number, and the traffic light is at the end of the red light or the beginning of the green light at the current moment, the corresponding risk level is high risk; When the comprehensive violation intensity index exceeds the preset threshold within a single time window but does not meet the continuous condition and / or the traffic light is in the yellow light switching stage, the corresponding risk level is medium risk; When the comprehensive violation intensity index does not exceed the preset threshold and the signal light is in the middle of the green light, the corresponding risk level is low risk.

5. The carbon emission management system for the transportation industry based on the Internet of Things according to claim 4, characterized in that, The classification of road segments into different risk levels also includes: Establish a communication connection with the traffic signal controller through the roadside IoT gateway to read the current phase status and remaining duration of the traffic light in real time; When the traffic light is in the red light state, the accumulated time from the time the red light is turned on to the current time and the preset total red light time are obtained synchronously; When the traffic light is in the green light state, the cumulative time from the time the green light is turned on to the current time and the preset remaining time of the green light are obtained synchronously.

6. The carbon emission management system for the transportation industry based on the Internet of Things according to claim 1, characterized in that The recommended paths with the lowest total risk value include: Upon receiving a motor vehicle navigation request, extracting a set of all feasible paths from a departure point to a destination; Eliminate routes that contain high-risk road segments and have no alternative detour solutions from the route set; For the remaining paths, the total risk value of the path is calculated according to the risk level of each road segment, where the high-risk road segment is assigned the maximum risk coefficient, the medium-risk road segment is assigned the middle risk coefficient, and the low-risk road segment is assigned the minimum risk coefficient; After arranging the total risk values of the paths in ascending order, the path with the lowest total risk value is recommended, and the risk level distribution on the path is marked with different colors in the electronic map interface.

7. The carbon emission management system for the transportation industry based on the Internet of Things according to claim 6, characterized in that The triggers for reassessing the risk level of the path include: The IoT sensing device detects that the density of real-time movement trajectories of non-motor vehicles in a certain road section of the current recommended route exceeds the historical level during the same period; Or it is detected that sudden traffic control occurs at the upstream intersection of the road segment, resulting in abnormal convergence of non-motor vehicle traffic; Or a group of non-motor vehicles are detected to be changing lanes and weaving in and out of the road continuously; When any trigger condition is met, the violation data within the most recent set time period is recalled for aggregation calculation, the comprehensive violation intensity index and risk level of the road segment are updated, and a new path recommendation instruction is generated based on the updated results.

8. The carbon emission management system for the transportation industry based on the Internet of Things according to claim 6, characterized in that, Real-time adjustments to the navigation path include: When the vehicle is driving along the recommended route, if the IoT sensing device detects that the risk level of the road ahead increases, the dynamic path switching mechanism is triggered; Recalculate the alternative path set based on the current position of the vehicle, and the alternative path set must meet the following conditions: exclude all newly promoted high-risk road segments, and the total length of the detour path does not exceed a set proportion of the original path length; Select a new path with the lowest total risk value from the set of alternative paths, and push a path change request to the driver through the vehicle terminal; If the driver confirms to accept the change, the navigation guidance trajectory is updated, and a steering instruction sequence is sent to the vehicle control system through the Internet of Vehicles. The steering instruction sequence includes the lane change prompts and steering angle suggestions required before each intersection; If the driver does not respond to the request, a warning signal will be broadcast to the vehicle behind through the Internet of Vehicles communication, triggering the vehicle behind to automatically maintain a safe distance. Based on the predicted motion trajectory of the non-motor vehicle in front, speed control instructions will be dynamically generated to guide the vehicle through high-risk road sections with a gentle acceleration and deceleration curve.

Images