Intersection anti-collision warning system and device based on intelligent transportation

Abstract

The present invention relates to the field of transportation technology, and more specifically to an intersection collision warning system and device based on intelligent transportation. The system determines the vehicle peak impact of the intersection within each preset second time period based on the number of vehicles passing through the intersection within the same preset second time period in all preset first time periods. The system then determines the comprehensive congestion performance of the intersection at the current moment by combining the number of vehicles traveling in each lane, the average vehicle speed, and the total duration of the green light. Furthermore, the system determines the collision risk index at the current moment by combining the difference in the number of pedestrians on the crosswalk at different times during the green light period, as well as the difference in the number of motor vehicles in adjacent lanes of the intersection at the same time, and performs intersection collision warning. The present invention improves the timeliness and accuracy of intersection collision warnings.

Description

Technical Field

[0001] The present invention relates to the field of traffic technology, and in particular to an intersection anti-collision warning system and device based on smart traffic. Background Art

[0002] Smart transportation leverages cutting-edge technologies such as the Internet of Things (IoT), cloud computing, big data, and artificial intelligence to enable real-time collection, analysis, and application of traffic information, thereby improving the operational efficiency and management of transportation systems. Intersection collision warning systems based on smart transportation are developing towards intelligent, collaborative, and automated approaches. Intersection collision warning is a technology used to improve intersection safety. It aims to predict potential collision risks based on complex intersection conditions and provide drivers with early warning signals, thereby preventing accidents.

[0003] Existing Problems: While existing intersection collision warning systems can identify routine traffic conditions to a certain extent, they often fail to accurately handle emergencies and irregular traffic behavior in complex and changing environments, such as the random behavior of pedestrians, the mixing of non-motorized vehicles and motor vehicles, and street vendors occupying the road. These uncertainties significantly increase the difficulty of predicting traffic flow dynamics, especially when multi-source data fusion is involved. Current methods suffer from delays, information loss, and errors. Data fusion between sensors often fails to respond to rapid changes in traffic conditions in a timely manner, resulting in an incomplete assessment of the intersection traffic situation and difficulty issuing accurate and effective warnings under the influence of multiple factors, thus reducing their practical application effectiveness in complex environments. Summary of the Invention

[0004] The present invention provides an intersection anti-collision warning system and device based on smart transportation to solve existing problems.

[0005] The present invention adopts the following technical solutions for the intersection anti-collision warning system and device based on intelligent transportation:

[0006] One embodiment of the present invention provides an intersection collision warning system based on smart transportation, which includes the following modules:

[0007] Traffic data collection module: used to obtain the number of vehicles passing through the intersection during each preset second time period within the preset first time period, and then obtain the number of vehicles traveling in each lane of the intersection, the average vehicle speed, the total length of time the green light is on, the number of motor vehicles at each moment, and the number of pedestrians on the crosswalk at each moment during the most recent green light period at the intersection during the current preset monitoring time period;

[0008] Peak period identification module: used to determine the vehicle peak impact of the intersection in each preset second time period according to the number of vehicles passing through the intersection in the same preset second time period in all preset first time periods;

[0009] Traffic congestion analysis module: used to determine the comprehensive congestion performance of the intersection at the current moment based on the number of vehicles traveling in each lane, the average vehicle speed, the total length of time the green light is on, and the impact of the vehicle peak;

[0010] Collision risk analysis module: used to determine the collision risk index at the current moment and perform intersection anti-collision warning based on the difference in the number of pedestrians on the crosswalk at different times during the green light period and the difference in the number of motor vehicles on adjacent lanes of the intersection at the same time, combined with the comprehensive congestion performance.

[0011] Furthermore, determining the vehicle peak impact degree of the intersection within each preset second time period includes:

[0012] In the In a preset first time period, obtaining the sum of the number of vehicles passing through the intersection in all preset second time periods, recording it as a first sum, and recording the ratio of the number of vehicles passing through the intersection in each preset second time period to the first sum as the ratio of the number of vehicles passing through the intersection in each preset second time period;

[0013] Get the The percentage of vehicles passing through the intersection in the second preset time period is reduced by The first and The normalized value of the mean difference of the proportion of vehicles passing through the intersection in the preset second time period is recorded as The combined impact of the peaks of the adjacent periods of the preset second time period;

[0014] Determine the peak coefficient of each preset second time period according to the proportion of the number of vehicles passing through the intersection in each preset second time period and the combined influence of the peak in the adjacent period;

[0015] Obtaining the product of the number of vehicles passing through the intersection in each preset second time period and the peak coefficient, and recording it as a first product;

[0016] In all preset first time periods, all The normalized value of the sum of the first products corresponding to the preset second time period is recorded as The impact of the vehicle peak at the intersection within a preset second time period.

[0017] Furthermore, determining the peak coefficient of each preset second time period includes:

[0018] The product of the proportion of the number of vehicles passing through the intersection in each preset second time period and the combined influence of the peak times in the adjacent time periods is recorded as the peak coefficient of each preset second time period.

[0019] Furthermore, determining the comprehensive congestion performance of the intersection at the current moment includes:

[0020] The ratio of the total green light duration of each lane at the intersection during the current preset monitoring time period to the duration of the current preset monitoring time period is recorded as the theoretical travel time ratio of each lane at the intersection during the current preset monitoring time period.

[0021] The inversely proportional normalized value of the product of the normalized value of the number of vehicles traveling in each lane of the intersection during the current preset monitoring time period and the average speed of the vehicles is recorded as the traffic obstruction degree of each lane of the intersection during the current preset monitoring time period;

[0022] Determine the comprehensive congestion performance of each lane at the intersection during the current preset monitoring time period based on the theoretical travel time proportion and traffic obstruction of each lane at the intersection during the current preset monitoring time period, combined with the vehicle peak impact;

[0023] The average of the comprehensive congestion performance of all lanes at the intersection during the current preset monitoring time period is recorded as the comprehensive congestion performance of the intersection at the current moment.

[0024] Furthermore, determining the comprehensive congestion performance of each lane at the intersection within the current preset monitoring time period includes:

[0025] Obtain the product of the theoretical travel time ratio and the traffic obstruction degree of each lane at the intersection during the current preset monitoring time period, and record it as the second product;

[0026] Obtain the product of the inverse proportional value of the theoretical travel time proportion of each lane at the intersection during the current preset monitoring time period and the vehicle peak influence degree at the intersection during the current preset second time period, recorded as the third product;

[0027] The sum of the second product and the third product is recorded as the comprehensive congestion performance of each lane at the intersection during the current preset monitoring time period.

[0028] Furthermore, determining the collision risk index at the current moment and performing an intersection collision warning includes:

[0029] Determine the estimated degree of forced pedestrian passage at the current moment based on the difference in the number of pedestrians on the crosswalk at different times during the most recent green light period at the intersection;

[0030] Determine the degree of visual obstruction at the current moment based on the difference in the number of motor vehicles in adjacent lanes of the intersection at the same moment within the current preset monitoring time period;

[0031] Determine the current collision risk index based on the comprehensive congestion performance of the intersection, the estimated degree of pedestrian forced passage, and the degree of visual obstruction at the current moment;

[0032] If the collision risk index at the current moment is less than or equal to the preset first threshold, it is determined to be low risk;

[0033] If the collision risk index at the current moment is greater than the preset first threshold and less than the preset second threshold, it is determined to be a medium risk;

[0034] If the collision risk index at the current moment is greater than or equal to the preset second threshold, it is determined to be a high risk.

[0035] Furthermore, determining the estimated degree of pedestrian forced passage at the current moment includes:

[0036] During the most recent green light period at the intersection, the moment when the number of pedestrians on the crosswalk is the largest is recorded as the marked time, and all moments after the marked time are recorded as reference times;

[0037] Obtain the normalized value of the time interval between the marked moment and each reference moment, and record it as the attention level of each reference moment;

[0038] The ratio of the number of pedestrians on the crosswalk at the marked time to the number of pedestrians on the crosswalk at each reference time is recorded as the peak proximity of the number of pedestrians at each reference time;

[0039] The product of the attention degree and the proximity to the peak number of pedestrians at each reference moment is obtained and recorded as the fourth product. The normalized value of the sum of the fourth products corresponding to all reference moments is recorded as the estimated degree of forced passage of pedestrians at the current moment.

[0040] Furthermore, determining the visual field obstruction degree at the current moment includes:

[0041] During the current preset monitoring time period, the absolute value of the difference in the number of motor vehicles between any two adjacent lanes of the intersection at the same time is obtained, and the maximum value of the absolute value of the difference in the number of motor vehicles between any two adjacent lanes of the intersection at all times is recorded as the maximum difference in the number of motor vehicles between any two adjacent lanes of the intersection, the time corresponding to the maximum difference in the number of motor vehicles between any two adjacent lanes of the intersection is recorded as the difference time of any two adjacent lanes of the intersection, the average of the maximum difference in the number of motor vehicles of all adjacent lanes of the intersection is recorded as the average difference in the number of motor vehicles in adjacent lanes, the normalized value of the standard deviation of the difference times of all adjacent lanes of the intersection is recorded as the adjacent lane field of view interference coefficient, and the normalized value of the product of the adjacent lane field of view interference coefficient and the average difference in the number of motor vehicles in adjacent lanes is recorded as the field of view obstruction at the current moment.

[0042] Furthermore, determining the collision risk index at the current moment based on the comprehensive congestion performance of the intersection, the pedestrian forced passage prediction degree, and the visual field obstruction degree at the current moment includes:

[0043] Obtain the product of the pedestrian's forced passage prediction degree at the current moment and the visual field obstruction degree at the current moment, which is recorded as the fifth product. Obtain the product of the inverse proportional value of the pedestrian's forced passage prediction degree at the current moment and the comprehensive congestion performance degree of the intersection at the current moment, which is recorded as the sixth product. The sum of the fifth product and the sixth product is recorded as the collision risk index at the current moment.

[0044] The present invention also proposes an intersection anti-collision warning device based on smart transportation, including a memory, a processor, and a computer program stored in the memory and runnable on the processor. The processor executes the computer program stored in the memory to implement the steps of the aforementioned intersection anti-collision warning system based on smart transportation.

[0045] The beneficial effects of this invention's technical solution include: accurately adapting to complex traffic conditions (such as uncontrolled pedestrian movement and mixed traffic of non-motorized vehicles) by integrating multi-source sensor data with dynamic analysis, updating the collision risk index in real time, and providing differentiated early warnings for vehicle-to-vehicle and vehicle-to-pedestrian collision risks based on peak hour patterns, lane congestion dynamics, and a quantitative model of visual obstruction. Furthermore, combined with graded warnings on electronic display screens, this significantly improves the timeliness and accuracy of early warnings while also alleviating traffic congestion and enhancing participants' risk avoidance awareness, providing a highly efficient and flexible solution for the coordinated optimization of safety and efficiency in intelligent transportation systems. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0047] Figure 1 This is a module flow chart of the intersection anti-collision warning system based on smart transportation of the present invention;

[0048] Figure 2 A schematic diagram of the intersection. DETAILED DESCRIPTION

[0049] To further illustrate the technical means and effects of the present invention to achieve the intended purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, describes in detail the specific implementation, structure, features, and effects of the intersection collision warning system and device based on smart transportation proposed by the present invention. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures, or characteristics of one or more embodiments may be combined in any suitable form.

[0050] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0051] The specific scheme of the intersection anti-collision warning system and device based on smart transportation provided by the present invention is described in detail below with reference to the accompanying drawings.

[0052] See also Figure 1 , which shows a module flow chart of an intersection anti-collision warning system based on smart transportation provided by one embodiment of the present invention. The system includes the following modules:

[0053] Module 101: Traffic data collection module.

[0054] This module is used to obtain the number of vehicles passing through the intersection in each preset second time period within the preset first time period, and then obtain the number of vehicles traveling in each lane of the intersection, the average vehicle speed, the total length of time the green light is on, the number of motor vehicles at each moment, and the number of pedestrians on the crosswalk at each moment during the most recent green light period at the intersection during the current preset monitoring time period.

[0055] In this embodiment, due to the complex conditions at intersections, basic traffic data at intersections, including vehicle and pedestrian counts, can be obtained using deployed intelligent transportation equipment. Vehicle counts are generated by using geomagnetic sensors, radar, or cameras (such as the YOLOv5 + DeepSort algorithm) to output real-time vehicle counts for each lane. Furthermore, using pedestrian re-identification (ReID) models or thermal imaging cameras, real-time counting of people at crosswalks or zebra crossings is achieved.

[0056] The prediction duration is the most recent week, that is, the week immediately before the current moment. The preset first time period is one day, the preset second time period is one hour, and the current preset monitoring time period is the most recent 300 seconds, that is, the most recent 300 seconds before the current moment. This is used as an example for description.

[0057] Thus, within the past week, the number of vehicles passing through the intersection in each preset second time period in the preset first time period is obtained, and then the number of vehicles traveling in each lane of the intersection, the average vehicle speed, the total length of time the green light is on, the number of motor vehicles at each moment, and the number of pedestrians on the crosswalk (all crosswalk areas of the intersection) at each moment during the most recent green light period at the intersection are obtained.

[0058] It should be noted that the number of motor vehicles in each lane at each moment in the intersection, as well as the number of pedestrians at each moment in the intersection's most recent green light period, are collected once per second. The average speed of vehicles in each lane of the intersection during the current preset monitoring period is calculated by taking the average speed of each vehicle passing through each lane of the intersection during the current preset monitoring period, and then taking the average of the average speeds of all vehicles passing through.

[0059] Module 102: Peak hour identification module.

[0060] The module is used to determine the vehicle peak impact of the intersection in each preset second time period based on the number of vehicles passing through the intersection in the same preset second time period in all preset first time periods.

[0061] It's important to note that daily commuting and school attendance create peak travel times on roads. During peak travel times, the number of vehicles on the road increases significantly, leading to higher traffic density. This is due to the concentration of shared traffic demand, particularly during weekday morning and evening rush hours or during certain holidays, when many people's travel purposes and times of day are highly consistent. Therefore, analyzing peak travel times at intersections will facilitate subsequent collision warning analysis.

[0062] Preferably, in one embodiment of the present invention, the method for obtaining the vehicle peak impact degree of the intersection in each preset second time period includes:

[0063] In the In a preset first time period, the sum of the number of vehicles passing through the intersection in all preset second time periods is obtained, recorded as the first sum value, and the ratio of the number of vehicles passing through the intersection in each preset second time period to the first sum value is recorded as the second sum value. The ratio of the number of vehicles passing through the intersection in each preset second time period in the preset first time period.

[0064] It should be noted that when analyzing peak traffic periods, the overall performance of multiple periods must be fully considered, especially the correlation between adjacent periods. If a single period has a high number of vehicles passing through, but the number of vehicles passing through adjacent periods does not show a similar growth or relatively consistent fluctuation trend, the peak performance of this single period should be considered a weak peak. Therefore, the trend of the peak period should be analyzed based on a comparison of a single period and adjacent periods, reflecting the comprehensive performance of multiple periods to understand the overall and lasting impact.

[0065] In the In the preset first time period, obtain the The percentage of vehicles passing through the intersection in the second preset time period is reduced by The first and The average of the differences in the proportion of vehicles passing through the intersection within the preset second time period The normalized value of The first time period of the preset The combined impact of the peaks of the adjacent time periods in the preset second time period.

[0066] In this embodiment, As The normalized value of is a linear normalization function used to normalize data values ​​to between 0 and 1. The second preset time period is When the first or last time period is the same as the first one, only one difference is calculated as .

[0067] The product of the proportion of the number of vehicles passing through the intersection in each preset second time period in each preset first time period and the combined influence of the peak times in the adjacent time periods is recorded as the peak coefficient of each preset second time period in each preset first time period.

[0068] It should be noted that: in this way, the peak coefficient of the intersection can be determined every day and every hour, and considering the long-term regularity of the peak period at the intersection, the peak impact of each period can be comprehensively determined based on the corresponding data of the same period of multiple days at the intersection in the recent period.

[0069] Obtain the product of the number of vehicles passing through the intersection and the peak coefficient in each preset second time period in each preset first time period, and record it as the first product. The sum of the first products corresponding to the preset second time period The normalized value of The impact of the vehicle peak at the intersection within a preset second time period.

[0070] Among them, As The normalized value of .

[0071] Module 103: Traffic congestion analysis module.

[0072] This module is used to determine the comprehensive congestion performance of the intersection at the current moment based on the number of vehicles traveling in each lane, the average vehicle speed, the total length of time the green light is on, and the vehicle peak impact.

[0073] It should be noted that during peak traffic hours, the traffic density on the road increases significantly, and there are many types of traffic participants, and the types of vehicles are complex (i.e., buses, taxis, and non-motorized vehicles), and the traffic flow is not uniform. As a result, drivers will frequently change lanes, causing further congestion on the road. At the same time, when there is traffic congestion, the following distance between vehicles will be closer, so the driver's reaction distance is relatively short in a complex traffic environment, especially when the vehicle in front suddenly slows down or stops, which can easily lead to rear-end collisions or other collisions. In this way, the congestion performance of the current real-time lane can be evaluated based on the number of vehicles and their speed. While considering the number of vehicles on the road, it is also necessary to take into account the speed of the vehicle. Because even if there are many vehicles on the road, the less the impact of factors such as frequent lane changes on the vehicle's speed, the lower the congestion performance of the real-time lane will be.

[0074] Preferably, in one embodiment of the present invention, the method for obtaining the comprehensive congestion performance of the intersection at the current moment includes:

[0075] The ratio of the total green light on time of each lane at the intersection during the current preset monitoring time period to the duration of the current preset monitoring time period is recorded as the theoretical travel time ratio of each lane at the intersection during the current preset monitoring time period.

[0076] The number of vehicles traveling in each lane of the intersection during the current preset monitoring time period The product of the normalized value of and the average speed of the vehicle The inversely proportional normalized value of is recorded as the traffic obstruction degree of each lane at the intersection during the current preset monitoring time period.

[0077] It should be noted that: in this embodiment, As The normalized value of As The inversely proportional normalized value. According to the number of vehicles traveling in each lane and the average speed of vehicles in the most recent monitoring period, the two are combined to determine the traffic obstruction of each lane. In this way, based on the vehicle peak impact of the time period of the current most recent monitoring period, and taking into account the theoretical travel time ratio of each lane, in the most recent monitoring period, the higher the theoretical travel time ratio of vehicles, the longer the green light time, and the easier it is to alleviate vehicle congestion. In this way, we can focus on the traffic obstruction of each lane (to reflect the current actual congestion performance). On the contrary, the lower the theoretical travel time ratio of vehicles, the more attention can be paid to the vehicle peak impact of the time period of the most recent monitoring period (that is, the chain reaction caused by the peak period), so as to comprehensively determine the comprehensive congestion performance of each lane in the most recent monitoring period.

[0078] Get the product of the theoretical travel time ratio of each lane at the intersection during the current preset monitoring time period and the traffic obstruction degree, recorded as the second product. Get the theoretical travel time ratio of each lane at the intersection during the current preset monitoring time period The product of the inverse proportional value of and the vehicle peak impact degree of the intersection in the preset second time period at the current moment is recorded as the third product, and the sum of the second product and the third product is recorded as the comprehensive congestion performance of each lane at the intersection in the current preset monitoring time period.

[0079] In this embodiment, , as The inverse proportional value of .

[0080] The average of the comprehensive congestion performance of all lanes at the intersection during the current preset monitoring time period is recorded as the comprehensive congestion performance of the intersection at the current moment.

[0081] Module 104: Collision risk analysis module.

[0082] This module is used to determine the collision risk index at the current moment and perform intersection anti-collision warning based on the difference in the number of pedestrians on the crosswalk at different times during the green light period, and the difference in the number of motor vehicles on adjacent lanes of the intersection at the same time, combined with the comprehensive congestion performance.

[0083] It's important to note that peak vehicle traffic often overlaps with peak travel times, meaning pedestrian demand also reaches peak levels. Furthermore, analysis can be conducted based on the number of people using crosswalks. If the number of people waiting at a crosswalk begins to pass as the green light turns on, and pedestrians continue to cross before and after the green light turns off, pedestrians may accelerate through traffic jams or attempt to force their way across the road in groups. This can significantly hinder vehicle traffic as the green light is about to turn on or has already turned on, increasing the risk of traffic accidents, particularly collisions between pedestrians and vehicles.

[0084] Preferably, in one embodiment of the present invention, the method for obtaining the collision risk index at the current moment includes:

[0085] During the most recent green light period at the intersection, the moment when the number of pedestrians on the crosswalk is the largest is recorded as the marked time, and the moments after the marked time are recorded as reference times.

[0086] It should be noted that if there are multiple times when the number of pedestrians at the crosswalk is the highest, the first time with the highest number of pedestrians is used. The number of pedestrians at the marked time is compared with the number of pedestrians at each subsequent time. The closer the number of pedestrians at each subsequent time is to the marked time, the greater the pedestrian demand during the most recent monitoring period. The crosswalk may show a "steady stream" of pedestrians at both ends. When the crosswalk light turns red, the number of people who have not yet fully crossed the crosswalk will increase, making it more likely that pedestrians will force their way through.

[0087] Get the time interval between the marked moment and each reference moment The normalized value of is recorded as the attention degree at each reference moment.

[0088] Among them, As The normalized value of .

[0089] The ratio of the number of pedestrians on the crosswalk at the marked moment to the number of pedestrians on the crosswalk at each reference moment is recorded as the proximity of the peak number of pedestrians at each reference moment.

[0090] Get the product of the attention degree and the proximity of the peak number of pedestrians at each reference moment, record it as the fourth product, and sum the fourth products corresponding to all reference moments The normalized value of is recorded as the estimated degree of pedestrian forced passage at the current moment.

[0091] Among them, As The normalized value of .

[0092] It should be noted that in the actual process of vehicle and pedestrian traffic, it will be affected by multiple factors, such as street vendors, which will affect the rapid passage of pedestrians due to their obstruction, and will also affect the vision of motor vehicle drivers, especially when there is a significant difference in the number of vehicles in adjacent lanes (such as when there are significantly more right-turning vehicles than vehicles in the straight lane). At this time, there will be a difference in vision, that is, when a pedestrian suddenly crosses the road in front of the right-turning vehicle, the driver will avoid it in time, while the vision of the driver in the straight lane is blocked by the right-turning vehicle (even when the green light is flashing), and he cannot detect the emergency in time, which may result in a higher collision risk. Figure 2 shown. Figure 2 If there's no traffic in the left-turn lane selected in the middle box, a car approaching from behind might accelerate if it finds the lane clear and the traffic light is flashing green. However, due to the obstruction of vision from vehicles in the lanes on either side, the driver might not be able to spot the pedestrian crossing the road in time, potentially causing a collision. Furthermore, if there are street vendors operating at the intersection, this not only affects pedestrians but also vehicles.

[0093] During the current preset monitoring time period, the absolute value of the difference in the number of motor vehicles between any two adjacent lanes of the intersection at the same time is obtained, and the maximum value of the absolute value of the difference in the number of motor vehicles between any two adjacent lanes of the intersection at all times is recorded as the maximum difference in the number of motor vehicles between any two adjacent lanes of the intersection, the time corresponding to the maximum difference in the number of motor vehicles between any two adjacent lanes of the intersection is recorded as the difference time of any two adjacent lanes of the intersection, the mean of the maximum difference in the number of motor vehicles of all adjacent lanes of the intersection is recorded as the average difference in the number of motor vehicles of adjacent lanes, and the standard deviation of the difference time of all adjacent lanes of the intersection is recorded as The normalized value of adjacent lane vision interference coefficient is recorded as the adjacent lane vision interference coefficient and the product of the average difference in the number of motor vehicles in adjacent lanes. The normalized value of is recorded as the visual obstruction at the current moment.

[0094] What needs to be explained is: As The normalized value of As The more scattered the differences between adjacent lanes are, the more significant differences in the number of lanes between them occur at different times. This results in more complex field of view obstruction during the most recent monitoring cycle (i.e., a single lane may have significantly fewer lanes than others, potentially obstructing the field of view in multiple directions).

[0095] It's important to note that collisions at intersections involve not only collisions between vehicles but also collisions between vehicles and pedestrians. A higher pedestrian forced-passing prediction score in the most recent monitoring cycle indicates greater focus on the degree of view obstruction in adjacent lanes (focusing on vehicle-pedestrian collisions). Conversely, a lower pedestrian forced-passing prediction score indicates greater focus on the intersection's multi-lane congestion performance (focusing on vehicle-to-vehicle collisions) during the current monitoring cycle.

[0096] Get the product of the pedestrian forced passage estimation degree at the current moment and the visual field obstruction degree at the current moment, record it as the fifth product, and set the pedestrian forced passage estimation degree at the current moment as The product of the inverse proportional value of and the comprehensive congestion performance of the intersection at the current moment is recorded as the sixth product, and the sum of the fifth product and the sixth product is recorded as the collision risk index at the current moment.

[0097] It should be noted that: in this embodiment, As The collision risk index is obtained every 5 minutes, which is described as an example.

[0098] The first threshold is preset to 0.35, and the second threshold is preset to 0.7, which is used as an example for description.

[0099] If the collision risk index at the current moment is less than or equal to the preset first threshold, it is determined to be a low risk.

[0100] If the collision risk index at the current moment is greater than the preset first threshold and less than the preset second threshold, it is determined to be a medium risk.

[0101] If the collision risk index at the current moment is greater than or equal to the preset second threshold, it is determined to be a high risk.

[0102] It should be noted that in this embodiment, the collision risk level of the intersection is displayed on an electronic display screen to provide a warning, thereby helping to reduce the occurrence of traffic accidents and improve traffic safety.

[0103] The present invention also provides an intersection anti-collision warning device based on smart transportation, including a memory, a processor, and a computer program stored in the memory and runnable on the processor. The processor executes the computer program stored in the memory to implement the steps of the aforementioned intersection anti-collision warning system based on smart transportation.

[0104] So far, the present invention is completed.

[0105] In summary, in an embodiment of the present invention, the vehicle peak impact degree of the intersection within each preset second time period is determined based on the number of vehicles passing through the intersection within the same preset second time period in all preset first time periods. The comprehensive congestion performance of the intersection at the current moment is determined by combining the number of vehicles traveling in each lane, the average vehicle speed, and the total green light duration. Furthermore, the collision risk index at the current moment is determined by combining the difference in the number of pedestrians on the crosswalk at different times during the green light period, as well as the difference in the number of motor vehicles in adjacent lanes of the intersection at the same time, and an intersection collision warning is performed. The present invention improves the timeliness and accuracy of intersection collision warnings.

[0106] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. The intersection anti-collision warning system based on intelligent transportation is characterized by: The system includes the following modules: Traffic data collection module: used to obtain the number of vehicles passing through the intersection during each preset second time period within the preset first time period, and then obtain the number of vehicles traveling in each lane of the intersection, the average vehicle speed, the total length of time the green light is on, the number of motor vehicles at each moment, and the number of pedestrians on the crosswalk at each moment during the most recent green light period at the intersection during the current preset monitoring time period; Peak period identification module: used to determine the vehicle peak impact of the intersection in each preset second time period according to the number of vehicles passing through the intersection in the same preset second time period in all preset first time periods; Traffic congestion analysis module: used to determine the comprehensive congestion performance of the intersection at the current moment based on the number of vehicles traveling in each lane, the average vehicle speed, the total length of time the green light is on, and the impact of the vehicle peak; Collision risk analysis module: used to determine the collision risk index at the current moment and perform intersection anti-collision warning based on the difference in the number of pedestrians on the crosswalk at different times during the green light period and the difference in the number of motor vehicles on adjacent lanes of the intersection at the same time, combined with the comprehensive congestion performance.

2. The intersection anti-collision warning system based on intelligent transportation according to claim 1 is characterized in that: Determining the impact of the vehicle peak at the intersection within each preset second time period includes: In the In a preset first time period, obtaining the sum of the number of vehicles passing through the intersection in all preset second time periods, recording it as a first sum, and recording the ratio of the number of vehicles passing through the intersection in each preset second time period to the first sum as the ratio of the number of vehicles passing through the intersection in each preset second time period; Get the The percentage of vehicles passing through the intersection in the second preset time period is reduced by The first and The normalized value of the mean difference of the proportion of vehicles passing through the intersection in the preset second time period is recorded as The combined impact of the peaks of the adjacent periods of the preset second time period; Determine the peak coefficient of each preset second time period according to the proportion of the number of vehicles passing through the intersection in each preset second time period and the combined influence of the peak in the adjacent period; Obtaining the product of the number of vehicles passing through the intersection in each preset second time period and the peak coefficient, and recording it as a first product; In all preset first time periods, all The normalized value of the sum of the first products corresponding to the preset second time period is recorded as The impact of the vehicle peak at the intersection within a preset second time period.

3. The intersection anti-collision warning system based on intelligent transportation according to claim 2 is characterized in that: Determining the peak coefficient of each preset second time period includes: The product of the proportion of the number of vehicles passing through the intersection in each preset second time period and the combined influence of the peak times in the adjacent time periods is recorded as the peak coefficient of each preset second time period.

4. The intersection anti-collision warning system based on intelligent transportation according to claim 1 is characterized in that: Determining the comprehensive congestion performance of the intersection at the current moment includes: The ratio of the total green light duration of each lane at the intersection during the current preset monitoring time period to the duration of the current preset monitoring time period is recorded as the theoretical travel time ratio of each lane at the intersection during the current preset monitoring time period. The inversely proportional normalized value of the product of the normalized value of the number of vehicles traveling in each lane of the intersection during the current preset monitoring time period and the average speed of the vehicles is recorded as the traffic obstruction degree of each lane of the intersection during the current preset monitoring time period; Determine the comprehensive congestion performance of each lane at the intersection during the current preset monitoring time period based on the theoretical travel time proportion and traffic obstruction of each lane at the intersection during the current preset monitoring time period, combined with the vehicle peak impact; The average of the comprehensive congestion performance of all lanes at the intersection during the current preset monitoring time period is recorded as the comprehensive congestion performance of the intersection at the current moment.

5. The intersection anti-collision warning system based on intelligent transportation according to claim 4 is characterized in that: Determining the comprehensive congestion performance of each lane at the intersection within the current preset monitoring time period includes: Obtain the product of the theoretical travel time ratio and the traffic obstruction degree of each lane at the intersection during the current preset monitoring time period, and record it as the second product; Obtain the product of the inverse proportional value of the theoretical travel time proportion of each lane at the intersection during the current preset monitoring time period and the vehicle peak influence degree at the intersection during the current preset second time period, recorded as the third product; The sum of the second product and the third product is recorded as the comprehensive congestion performance of each lane at the intersection during the current preset monitoring time period.

6. The intersection anti-collision warning system based on intelligent transportation according to claim 1 is characterized in that: Determining the collision risk index at the current moment and performing an intersection collision warning includes: Determine the estimated degree of forced pedestrian passage at the current moment based on the difference in the number of pedestrians on the crosswalk at different times during the most recent green light period at the intersection; Determine the degree of visual obstruction at the current moment based on the difference in the number of motor vehicles in adjacent lanes of the intersection at the same moment within the current preset monitoring time period; Determine the current collision risk index based on the comprehensive congestion performance of the intersection, the estimated degree of pedestrian forced passage, and the degree of visual obstruction at the current moment; If the collision risk index at the current moment is less than or equal to the preset first threshold, it is determined to be low risk; If the collision risk index at the current moment is greater than the preset first threshold and less than the preset second threshold, it is determined to be a medium risk; If the collision risk index at the current moment is greater than or equal to the preset second threshold, it is determined to be a high risk.

7. The intersection anti-collision warning system based on intelligent transportation according to claim 6 is characterized in that: Determining the estimated degree of pedestrian forced passage at the current moment includes: During the most recent green light period at the intersection, the moment when the number of pedestrians on the crosswalk is the largest is recorded as the marked time, and all moments after the marked time are recorded as reference times; Obtain the normalized value of the time interval between the marked moment and each reference moment, and record it as the attention level of each reference moment; The ratio of the number of pedestrians on the crosswalk at the marked time to the number of pedestrians on the crosswalk at each reference time is recorded as the peak proximity of the number of pedestrians at each reference time; The product of the attention degree and the proximity to the peak number of pedestrians at each reference moment is obtained and recorded as the fourth product. The normalized value of the sum of the fourth products corresponding to all reference moments is recorded as the estimated degree of forced passage of pedestrians at the current moment.

8. The intersection anti-collision warning system based on intelligent transportation according to claim 6 is characterized in that: Determining the visual field obstruction at the current moment includes: During the current preset monitoring time period, the absolute value of the difference in the number of motor vehicles between any two adjacent lanes of the intersection at the same time is obtained, and the maximum value of the absolute value of the difference in the number of motor vehicles between any two adjacent lanes of the intersection at all times is recorded as the maximum difference in the number of motor vehicles between any two adjacent lanes of the intersection, the time corresponding to the maximum difference in the number of motor vehicles between any two adjacent lanes of the intersection is recorded as the difference time of any two adjacent lanes of the intersection, the average of the maximum difference in the number of motor vehicles of all adjacent lanes of the intersection is recorded as the average difference in the number of motor vehicles in adjacent lanes, the normalized value of the standard deviation of the difference times of all adjacent lanes of the intersection is recorded as the adjacent lane field of view interference coefficient, and the normalized value of the product of the adjacent lane field of view interference coefficient and the average difference in the number of motor vehicles in adjacent lanes is recorded as the field of view obstruction at the current moment.

9. The intersection anti-collision warning system based on intelligent transportation according to claim 6 is characterized in that: Determining the collision risk index at the current moment based on the comprehensive congestion performance of the intersection, the pedestrian forced passage estimation degree, and the visual obstruction degree at the current moment includes: Obtain the product of the pedestrian's forced passage prediction degree at the current moment and the visual field obstruction degree at the current moment, which is recorded as the fifth product. Obtain the product of the inverse proportional value of the pedestrian's forced passage prediction degree at the current moment and the comprehensive congestion performance degree of the intersection at the current moment, which is recorded as the sixth product. The sum of the fifth product and the sixth product is recorded as the collision risk index at the current moment.

10. An intersection anti-collision warning device based on intelligent transportation, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the computer program is executed by a processor, the steps of the intersection anti-collision warning system based on smart transportation as described in any one of claims 1 to 9 are implemented.

Images