Prediction method for minimum driving speed interval of vehicles at intersection

Abstract

The invention is suitable for the technical field of intelligent traffic, and relates to an intersection vehicle minimum driving speed interval prediction method comprising the following steps: S10, obtaining vehicle and intersection scene information based on a camera and a 3D laser sensor; s20, acquiring behavior characteristics and dynamic characteristics based on the car following behavior model; s30, establishing a hierarchical structure for predicting the minimum speed interval of the target vehicle driving from the current position to the intersection; s40, establishing a five-level residual neural network for each layer of model of the hierarchical structure, and obtaining the possibility that the minimum speed of the target vehicle is in each interval; and S50, the intersection speed guiding system selects a suitable minimum speed prediction interval according to a preset possibility threshold value so as to improve the reliability and accuracy of the intersection-crossing speed guiding system. The method is simple in process and convenient to operate, realizes prediction of the driving minimum speed interval of the vehicle driving at the intersection from the current position to the intersection, is accurate in prediction result, and improves the traffic communication efficiency of the intersection.

Description

Technical Field

[0001] The invention belongs to the field of intelligent transportation technology, and in particular relates to a method for predicting a minimum driving speed interval of vehicles at an intersection. Background Art

[0002] With the rapid development of vehicle-road cooperative driving systems, the demand for various intelligent driving assistance systems and automatic driving systems is gradually increasing. By obtaining the status information of the current vehicle, other vehicles in the current lane and traffic lights, the driving strategy of the current vehicle can be effectively optimized, and the comfort and energy efficiency of vehicle driving can be improved. By predicting the minimum speed of vehicles traveling at intersections over a period of time, it can provide the most critical influencing factors for optimizing the target driving speed of the vehicle and optimizing the lane selection strategy of the vehicle. Further combining micro-traffic flow and intersection information is also conducive to optimizing the timing strategy of traffic lights and improving traffic efficiency at intersections.

[0003] In fact, the technical solutions recorded in the prior art are only to predict the time when a vehicle arrives at an intersection. If there are other vehicles in front of the target vehicle, it is necessary to recursively predict the arrival time of the front vehicle and calculate the time when the vehicle arrives at the intersection based on the arrival time of the front vehicle and the distance and relative speed between the target vehicle and the front vehicle. For example, the patent with announcement number CN111681413B provides a method and device for real-time prediction of the time when a motor vehicle passes through a signal-controlled intersection. The method includes: obtaining driving state information of a motor vehicle traveling on a section of a signal-controlled intersection, including current driving position information and current speed information, and obtaining signal light information set at the signal-controlled intersection, judging whether there are other motor vehicles traveling in front of the currently predicted motor vehicle and whether there are signal lights in front. When there are motor vehicles traveling in front of the predicted motor vehicle and there are no signal lights, judging whether the speed of the motor vehicle in front is zero. When it is not zero, determining the time when the currently predicted motor vehicle arrives at the predicted position based on the distance between the currently predicted motor vehicle and the motor vehicle in front, the speed of the currently predicted motor vehicle, and the time when the motor vehicle in front of the currently predicted motor vehicle arrives at the predicted position. The prediction process of this patent is complex and inefficient, and it only predicts the time it takes for a motor vehicle to arrive at a predicted location, but does not record a technical solution for calculating the minimum driving speed of a motor vehicle.

[0004] Therefore, how to accurately and efficiently predict the minimum speed of vehicles traveling at an intersection over a period of time to improve the traffic efficiency of the intersection is an urgent problem to be solved by personnel in this technical field. Summary of the invention

[0005] In view of the deficiencies of the prior art, the purpose of the present invention is to provide a method for predicting the minimum driving speed range of vehicles at an intersection, so as to solve the problem that the prior art cannot accurately predict the minimum driving speed range of vehicles at an intersection, resulting in low traffic efficiency.

[0006] In order to solve the above technical problems, the present invention adopts the following technical solutions:

[0007] The present invention provides a method for predicting the minimum speed range of vehicles at an intersection, comprising the following steps:

[0008] S10, obtaining vehicle and intersection scene information based on cameras and 3D laser sensors;

[0009] S20, obtaining behavior characteristics and dynamic characteristics based on the vehicle-following behavior model;

[0010] S30, establishing a hierarchical structure for predicting the minimum speed interval of the target vehicle traveling from the current position to the intersection;

[0011] S40, establishing a five-level residual neural network for each layer model of the hierarchical structure, and obtaining the possibility that the minimum speed of the target vehicle is in each interval;

[0012] S50. The intersection speed guidance system selects a suitable minimum speed prediction interval according to a predetermined possibility threshold value, so as to improve the reliability and accuracy of the intersection speed guidance system.

[0013] Furthermore, the specific steps of step S10 are as follows:

[0014] S101, install cameras and 3D laser sensors on street lamps;

[0015] S102, identifying and segmenting the vehicle type, vehicle color, vehicle position and two-dimensional enclosing shape in the image based on the traffic light camera and the vehicle recognition algorithm;

[0016] S103. By installing a traffic light camera near a 3D laser sensor, all 3D point clouds of the intersection traffic scene are obtained. The 3D point cloud of each vehicle at the intersection is obtained by matching the vehicle camera and the 3D laser sensor, thereby obtaining the vehicle's wheelbase information, speed information, three-dimensional shape information, and distance information from the intersection.

[0017] Furthermore, the specific steps of step S20 are as follows:

[0018] S201, based on the collected data for a period of time and data analysis, obtain the type, color, position, two-dimensional surrounding shape, wheelbase information, speed information, three-dimensional shape information, distance information from the intersection, acceleration information and deceleration information of each vehicle at each moment at the intersection;

[0019] S202, based on the collected position information of each vehicle at the intersection at each moment, parameter estimation is performed on the following behavior model of vehicles of different vehicle types, different axle lengths, and different queue positions;

[0020] S203, based on the following behavior model of each vehicle after parameter estimation and the intersection scene information, multiple simulations are performed to obtain the dynamic characteristics of the target vehicle in the process of reaching the intersection.

[0021] Furthermore, in step S202, the parameters of the vehicle following behavior model include: maximum acceleration, maximum deceleration, vehicle following target distance, vehicle following target headway, parking target distance, maximum driving speed, driver reaction time, and vehicle start and stop time.

[0022] Furthermore, in step S203, the dynamic features include the average, maximum, minimum, 25% percentile, 50% percentile, and 75% percentile of the speed and time of the target vehicle to the intersection, and the average, maximum, minimum, 25% percentile, 50% percentile, and 75% percentile of the target vehicle at the minimum speed and time.

[0023] Furthermore, the specific steps of step S30 are as follows:

[0024] S301, dividing the minimum speed of the target vehicle when it is driving to the intersection into 9 intervals and assigning corresponding category labels;

[0025] S302. According to the congestion definition and data analysis, two adjacent speed intervals are manually merged each time, and finally a hierarchy of minimum speed intervals is formed.

[0026] Furthermore, in step S301, the 9 minimum speed intervals are [0, 5km / h), [5km / h.10km / h), [10km / h.15km / h), [15km / h, 20km / h), [20km / h, 25km / h), [25km / h, 30km / h), [35km / h, 40km / h), [40km / h, 45km / h) and [45km / h, 120km / h).

[0027] Furthermore, in step S302, the hierarchy of the minimum speed interval is as follows:

[0028] The minimum speed intervals of the 9th layer are [0,5km / h), [5km / h.10km / h), [10km / h.15km / h), [15km / h,20km / h), [20km / h,25km / h), [25km / h,30km / h), [35km / h,40km / h), [40km / h,45km / h) and [45km / h,120km / h];

[0029] The minimum speed intervals of the 8th layer are [0, 5km / h), [5km / h, 15km / h), [15km / h, 20km / h), [20km / h, 25km / h), [25km / h, 30km / h), [35km / h, 40km / h), [40km / h, 45km / h) and [45km / h, 120km / h];

[0030] The minimum speed intervals for the 7th layer are [0, 5km / h), [5km / h, 20km / h), [20km / h, 25km / h), [25km / h, 30km / h), [35km / h, 40km / h), [40km / h, 45km / h) and [45km / h, 120km / h];

[0031] The minimum speed intervals for the sixth layer are [0, 5km / h), [5km / h, 20km / h), [20km / h, 25km / h), [25km / h, 30km / h), [35km / h, 40km / h) and [40km / h, 120km / h];

[0032] The minimum speed intervals for the 5th layer are [0,5km / h), [5km / h, 20km / h), [20km / h, 25km / h), [25km / h, 30km / h) and [35km / h, 120km / h];

[0033] The minimum speed intervals for the 4th layer are [0,5km / h), [5km / h,25km / h), [25km / h,30km / h) and [35km / h,120km / h];

[0034] The minimum speed intervals for the third layer are [0,5km / h), [5km / h,30km / h) and [35km / h,120km / h];

[0035] The minimum speed intervals of the second layer are [0,5km / h) and [5km / h,120km / h];

[0036] The minimum speed interval of the first layer is: [0,120km / h].

[0037] Furthermore, the specific steps of step S40 are as follows:

[0038] S401. Based on the hierarchical structure of the minimum speed interval, a 5-level residual neural network is designed and trained for each layer from layer 2 to layer 9. Each level of the network has 128 nodes, and the last level is the output level, which is of softMax type.

[0039] S402, by collecting the status and feature data of vehicles queuing at the intersection, a data set is established to train eight hierarchical neural networks;

[0040] S403, predicting the minimum speed range and corresponding probability of the target vehicle in the process of reaching the intersection based on 8 neural networks.

[0041] Furthermore, in step S402, the neural network corresponding to the 9th layer model is fully trained during the training process, and the neural network corresponding to the 8th to 2nd layer models only trains the last level of the neural network, and the other levels of neural networks reuse the neural network corresponding to the 9th layer model.

[0042] Compared with the prior art, the method for predicting the minimum speed interval of vehicles at an intersection provided by the present invention has at least the following beneficial effects:

[0043] The technical solutions recorded in the prior art only predict the time when the vehicle arrives at the intersection, and only realize the prediction of the time when the motor vehicle arrives at the predicted position, but do not record the technical solutions for calculating the minimum driving speed of the motor vehicle, and the prediction process is complicated and inefficient. The present invention can more accurately obtain the driving status and traffic status of each vehicle at the intersection based on the traffic light camera and 3D laser sensor placed on the street lamp; the present invention is based on the driving status and traffic status of the vehicle, so as to more accurately estimate the following vehicle dynamic model of each driving behavior; the prediction model of the present invention not only combines the static characteristics of the scene but also adds various behavioral characteristics of the following vehicle behavior model based on data fitting and dynamic characteristics in multiple simulations, so as to improve the accuracy of predicting the minimum speed interval in the process of driving to the intersection; the present patent establishes a hierarchical minimum speed prediction model from coarse to fine, and each layer uses a 5-level neural network with the same structure to predict the possibility of the minimum speed of the target vehicle in the process of driving to the intersection in each minimum speed interval at the current level, which effectively expands the prediction range, improves the accuracy of the prediction, and selects a suitable minimum speed prediction interval according to a predetermined possibility threshold, thereby improving the traffic efficiency of the intersection. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] In order to more clearly illustrate the scheme of the present invention, a brief introduction is given below to the figures required for use in the description of the embodiments. Obviously, the figures described below are some embodiments of the present invention. For ordinary technicians in this field, other figures can be obtained based on these figures without paying any creative work.

[0045] Figure 1 A flowchart of a method for predicting the minimum speed range of vehicles at an intersection provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0046] In order to facilitate the understanding of the present invention, the present invention will be described more fully below with reference to the relevant drawings. The preferred embodiments of the present invention are given in the drawings. However, the present invention can be implemented in many different forms and is not limited to the embodiments described herein. On the contrary, the purpose of providing these embodiments is to make the understanding of the disclosure of the present invention more thorough and comprehensive.

[0047] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art of the present invention. The terms used in the specification of the present invention herein are only for the purpose of describing specific embodiments and are not intended to limit the present invention.

[0048] The present invention provides a method for predicting the minimum speed interval of vehicles at an intersection, which is applied to predicting the minimum speed interval of vehicles within a period of time. The method for predicting the minimum speed interval of vehicles at an intersection comprises the following steps:

[0049] S10, obtaining vehicle and intersection scene information based on cameras and 3D laser sensors;

[0050] S20, obtaining behavior characteristics and dynamic characteristics based on the vehicle-following behavior model;

[0051] S30, establishing a hierarchical structure for predicting the minimum speed interval of the target vehicle traveling from the current position to the intersection;

[0052] S40, establishing a five-level residual neural network for each layer model of the hierarchical structure, and obtaining the possibility that the minimum speed of the target vehicle is in each interval;

[0053] S50. The intersection speed guidance system selects a suitable minimum speed prediction interval according to a predetermined possibility threshold value, so as to improve the reliability and accuracy of the intersection speed guidance system.

[0054] The invention has a simple process and convenient operation, and realizes the prediction of the minimum speed interval of vehicles traveling at the intersection from the current position to the intersection. The prediction result is accurate, and the traffic communication efficiency of the intersection is improved.

[0055] In order to enable those skilled in the art to better understand the solutions of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings.

[0056] The present invention provides a method for predicting the minimum speed interval of vehicles at an intersection, which is applied to predicting the minimum speed interval of vehicles within a period of time, such as Figure 1 As shown, in this embodiment, the method for predicting the minimum driving speed interval of vehicles at the intersection includes the following steps:

[0057] S10, obtaining vehicle and intersection scene information based on cameras and 3D laser sensors;

[0058] Specifically, the specific steps of step S10 are as follows:

[0059] S101, install cameras and 3D laser sensors on street lamps;

[0060] S102, identifying and segmenting the vehicle type, vehicle color, vehicle position and two-dimensional enclosing shape in the image based on the traffic light camera and the vehicle recognition algorithm;

[0061] S103. By installing a traffic light camera near a 3D laser sensor, all 3D point clouds of the traffic scene at the intersection are obtained. The 3D point cloud of each vehicle at the intersection is obtained through the matching algorithm between the vehicle camera and the 3D laser sensor, thereby obtaining the vehicle's wheelbase information, speed information, three-dimensional shape information, and distance information from the intersection, so as to more accurately predict the following behavior model of each vehicle.

[0062] S20, obtaining behavior characteristics and dynamic characteristics based on the vehicle-following behavior model;

[0063] Specifically, the specific steps of step S20 are as follows:

[0064] S201, based on the collected data for a period of time and data analysis, obtain the type, color, position, two-dimensional surrounding shape, wheelbase information, speed information, three-dimensional shape information, distance information from the intersection, acceleration information and deceleration information of each vehicle at each moment at the intersection;

[0065] S202, based on the collected position information of each vehicle at the intersection at each moment, parameter estimation is performed on the following behavior model of vehicles of different vehicle types, different axle lengths, and different queue positions;

[0066] Further, in this embodiment, the parameters of the vehicle following behavior model include: maximum acceleration, maximum deceleration, vehicle following target distance, vehicle following target headway, parking target distance, maximum driving speed, driver reaction time, vehicle start and stop time;

[0067] S203, based on the following behavior model of each vehicle after parameter estimation and the intersection scene information, multiple simulations are performed to obtain the dynamic characteristics of the target vehicle in the process of reaching the intersection;

[0068] Furthermore, in this embodiment, the dynamic features include the average, maximum, minimum, 25% percentile, 50% percentile, and 75% percentile of the speed and time of the target vehicle to the intersection, the average, maximum, minimum, 25% percentile, 50% percentile, and 75% percentile of the target vehicle at the minimum speed and time, and are added to the input features of the neural network. The model input features of the embodiment of the present invention not only include the static features of the intersection scene, but also add the following vehicle behavior model features based on data fitting, and the dynamic features of the target vehicle reaching the intersection after multiple simulations of the following vehicle behavior model. The output value of the neural network is the possibility that the minimum speed of the target vehicle in the process of driving to the intersection is in each minimum speed interval, which improves the accuracy of predicting the minimum speed interval in the process of driving to the intersection.

[0069] S30, establishing a hierarchical structure for predicting the minimum speed interval of the target vehicle traveling from the current position to the intersection;

[0070] Specifically, the specific steps of step S30 are as follows:

[0071] S301, dividing the minimum speed of the target vehicle when it is driving to the intersection into 9 intervals and assigning corresponding category labels;

[0072] Furthermore, in the present embodiment, the nine intervals of minimum speed are respectively [0, 5km / h), [5km / h.10km / h), [10km / h.15km / h), [15km / h, 20km / h), [20km / h, 25km / h), [25km / h, 30km / h), [35km / h, 40km / h), [40km / h, 45km / h) and [45km / h, 120km / h). The embodiment of the present invention adopts a coarse-to-fine hierarchical prediction method for the interval of the minimum speed of the vehicle. By expanding the prediction range and improving the accuracy of the prediction, the driving assistance system can select a suitable minimum speed prediction interval according to a predetermined possibility threshold (such as greater than 95%), give a suitable guiding speed for the intersection, and improve the reliability and accuracy of a driving decision system such as the intersection speed guidance system.

[0073] S302. According to the congestion definition and data analysis, two adjacent speed intervals are manually merged each time, and finally a hierarchy of minimum speed intervals is formed.

[0074] Specifically, in this embodiment, the hierarchical structure of the minimum speed interval is as follows:

[0075] The minimum speed intervals of the 9th layer are [0,5km / h), [5km / h.10km / h), [10km / h.15km / h), [15km / h,20km / h), [20km / h,25km / h), [25km / h,30km / h), [35km / h,40km / h), [40km / h,45km / h) and [45km / h,120km / h];

[0076] The minimum speed intervals of the 8th layer are [0, 5km / h), [5km / h, 15km / h), [15km / h, 20km / h), [20km / h, 25km / h), [25km / h, 30km / h), [35km / h, 40km / h), [40km / h, 45km / h) and [45km / h, 120km / h];

[0077] The minimum speed intervals for the 7th layer are [0, 5km / h), [5km / h, 20km / h), [20km / h, 25km / h), [25km / h, 30km / h), [35km / h, 40km / h), [40km / h, 45km / h) and [45km / h, 120km / h];

[0078] The minimum speed intervals for the sixth layer are [0, 5km / h), [5km / h, 20km / h), [20km / h, 25km / h), [25km / h, 30km / h), [35km / h, 40km / h) and [40km / h, 120km / h];

[0079] The minimum speed intervals for the 5th layer are [0,5km / h), [5km / h, 20km / h), [20km / h, 25km / h), [25km / h, 30km / h) and [35km / h, 120km / h];

[0080] The minimum speed intervals for the 4th layer are [0,5km / h), [5km / h,25km / h), [25km / h,30km / h) and [35km / h,120km / h];

[0081] The minimum speed intervals for the third layer are [0,5km / h), [5km / h,30km / h) and [35km / h,120km / h];

[0082] The minimum speed intervals of the second layer are [0,5km / h) and [5km / h,120km / h];

[0083] The minimum speed interval of the first layer is: [0,120km / h].

[0084] S40, establishing a five-level residual neural network for each layer model of the hierarchical structure, and obtaining the possibility that the minimum speed of the target vehicle is in each interval;

[0085] Specifically, the specific steps of step S40 are as follows:

[0086] S401. Based on the hierarchical structure of the minimum speed interval, a 5-level residual neural network is designed and trained for each layer from layer 2 to layer 9. Each level of the network has 128 nodes, and the last level is the output level, which is of softMax type.

[0087] S402, by collecting the status and feature data of vehicles queuing at the intersection, a data set is established to train eight hierarchical neural networks;

[0088] Furthermore, in this embodiment, the neural network corresponding to the 9th layer model is fully trained during the training process, and the neural network corresponding to the 8th to 2nd layer models only trains the last level of the neural network, and the neural networks of other levels reuse the neural network corresponding to the 9th layer model, thereby reducing the complexity of building and training the neural network;

[0089] S403, predicting the minimum speed range and corresponding probability of the target vehicle in the process of reaching the intersection based on 8 neural networks.

[0090] S50. The intersection speed guidance system selects a suitable minimum speed prediction interval according to a predetermined possibility threshold value, so as to improve the reliability and accuracy of the intersection speed guidance system.

[0091] Compared with the prior art, the method for predicting the minimum driving speed interval of vehicles at intersections described in the above embodiments only predicts the time when the vehicle arrives at the intersection, and only realizes the time when the motor vehicle arrives at the predicted position, but does not record the technical solution for calculating the minimum driving speed of the motor vehicle, and the prediction process is complicated and inefficient. The present invention can more accurately obtain the driving state and traffic state of each vehicle at the intersection based on the traffic light camera and 3D laser sensor placed on the street lamp; the present invention is based on the driving state and traffic state of the vehicle, so as to more accurately estimate the following vehicle dynamic model of each driving behavior; the prediction model of the present invention not only combines the static characteristics of the scene but also adds various behavioral characteristics of the following vehicle behavior model based on data fitting and dynamic characteristics in multiple simulations, so as to improve the accuracy of predicting the minimum speed interval in the process of driving to the intersection; the present patent establishes a hierarchical minimum speed prediction model from coarse to fine, and each layer uses a 5-level neural network with the same structure to predict the possibility of the minimum speed of the target vehicle in the process of driving to the intersection in each minimum speed interval at the current level, which effectively expands the prediction range, improves the accuracy of the prediction, and selects a suitable minimum speed prediction interval according to a predetermined possibility threshold, thereby improving the traffic efficiency of the intersection.

[0092] Obviously, the embodiments described above are only preferred embodiments of the present invention, rather than all embodiments. The preferred embodiments of the present invention are shown in the accompanying drawings, but they do not limit the patent scope of the present invention. The present invention can be implemented in many different forms. On the contrary, the purpose of providing these embodiments is to make the understanding of the disclosure of the present invention more thorough and comprehensive. Although the present invention has been described in detail with reference to the aforementioned embodiments, for those skilled in the art, it is still possible to modify the technical solutions recorded in the aforementioned specific embodiments, or to perform equivalent replacements for some of the technical features therein. Any equivalent structure made using the contents of the specification and drawings of the present invention, directly or indirectly used in other related technical fields, is also within the scope of patent protection of the present invention.

Claims

1. A method for predicting the minimum speed interval of vehicles at an intersection, characterized in that: The following steps are involved: S10, obtaining vehicle and intersection scene information based on cameras and 3D laser sensors; S20, obtaining behavior characteristics and dynamic characteristics based on the vehicle-following behavior model; S30, establishing a hierarchical structure for predicting the minimum speed interval of the target vehicle traveling from the current position to the intersection; S40, establishing a five-level residual neural network for each layer model of the hierarchical structure, and obtaining the possibility that the minimum speed of the target vehicle is in each interval; S50. The intersection speed guidance system selects a suitable minimum speed prediction interval according to a predetermined possibility threshold value, so as to improve the reliability and accuracy of the intersection speed guidance system.

2. The method for predicting the minimum speed interval of vehicles at an intersection according to claim 1, characterized in that: The specific steps of step S10 are as follows: S101, install cameras and 3D laser sensors on street lamps; S102, identifying and segmenting the vehicle type, vehicle color, vehicle position and two-dimensional enclosing shape in the image based on the traffic light camera and the vehicle recognition algorithm; S103. By installing a traffic light camera near a 3D laser sensor, all 3D point clouds of the intersection traffic scene are obtained. The 3D point cloud of each vehicle at the intersection is obtained by matching the vehicle camera and the 3D laser sensor, thereby obtaining the vehicle's wheelbase information, speed information, three-dimensional shape information, and distance information from the intersection.

3. The method for predicting the minimum speed interval of vehicles at an intersection according to claim 1, characterized in that: The specific steps of step S20 are as follows: S201, based on the collected data for a period of time and data analysis, obtain the type, color, position, two-dimensional surrounding shape, wheelbase information, speed information, three-dimensional shape information, distance information from the intersection, acceleration information and deceleration information of each vehicle at each moment at the intersection; S202, based on the collected position information of each vehicle at the intersection at each moment, parameter estimation is performed on the following behavior model of vehicles of different vehicle types, different axle lengths, and different queue positions; S203, based on the following behavior model of each vehicle after parameter estimation and the intersection scene information, multiple simulations are performed to obtain the dynamic characteristics of the target vehicle in the process of reaching the intersection.

4. The method for predicting the minimum speed interval of vehicles at an intersection according to claim 3, characterized in that: In step S202, the parameters of the vehicle following behavior model include: maximum acceleration, maximum deceleration, target following distance, target following headway, target stopping distance, maximum driving speed, driver reaction time, and vehicle start and stop time.

5. The method for predicting the minimum speed interval of vehicles at an intersection according to claim 3, characterized in that: In step S203, the dynamic features include the average value, maximum value, minimum value, 25% percentile value, 50% percentile value, 75% percentile value of the speed and time of the target vehicle to the intersection, and the average value, maximum value, minimum value, 25% percentile value, 50% percentile value, 75% percentile value of the target vehicle at the minimum speed and time.

6. The method for predicting the minimum speed interval of vehicles at an intersection according to claim 1, characterized in that: The specific steps of step S30 are as follows: S301, dividing the minimum speed of the target vehicle when it is driving to the intersection into 9 intervals and assigning corresponding category labels; S302. According to the congestion definition and data analysis, two adjacent speed intervals are manually merged each time, and finally a hierarchy of minimum speed intervals is formed.

7. The method for predicting the minimum speed interval of vehicles at an intersection according to claim 6, characterized in that: In step S301, the 9 minimum speed intervals are [0, 5km / h), [5km / h.10km / h), [10km / h.15km / h), [15km / h, 20km / h), [20km / h, 25km / h), [25km / h, 30km / h), [35km / h, 40km / h), [40km / h, 45km / h) and [45km / h, 120km / h).

8. The method for predicting the minimum speed interval of vehicles at an intersection according to claim 6, characterized in that: In step S302, the hierarchy of the minimum speed interval is as follows: The minimum speed intervals of the 9th layer are [0,5km / h), [5km / h.10km / h), [10km / h.15km / h), [15km / h,20km / h), [20km / h,25km / h), [25km / h,30km / h), [35km / h,40km / h), [40km / h,45km / h) and [45km / h,120km / h]; The minimum speed intervals of the 8th layer are [0, 5km / h), [5km / h, 15km / h), [15km / h, 20km / h), [20km / h, 25km / h), [25km / h, 30km / h), [35km / h, 40km / h), [40km / h, 45km / h) and [45km / h, 120km / h]; The minimum speed intervals for the 7th layer are [0, 5km / h), [5km / h, 20km / h), [20km / h, 25km / h), [25km / h, 30km / h), [35km / h, 40km / h), [40km / h, 45km / h) and [45km / h, 120km / h]; The minimum speed intervals for the sixth layer are [0, 5km / h), [5km / h, 20km / h), [20km / h, 25km / h), [25km / h, 30km / h), [35km / h, 40km / h) and [40km / h, 120km / h]; The minimum speed intervals for the 5th layer are [0,5km / h), [5km / h, 20km / h), [20km / h, 25km / h), [25km / h, 30km / h) and [35km / h, 120km / h]; The minimum speed intervals for the 4th layer are [0,5km / h), [5km / h,25km / h), [25km / h,30km / h) and [35km / h,120km / h]; The minimum speed intervals for the third layer are [0,5km / h), [5km / h,30km / h) and [35km / h,120km / h]; The minimum speed intervals of the second layer are [0,5km / h) and [5km / h,120km / h]; The minimum speed interval of the first layer is: [0,120km / h].

9. The method for predicting the minimum speed interval of vehicles at an intersection according to claim 8, characterized in that: The specific steps of step S40 are as follows: S401. Based on the hierarchical structure of the minimum speed interval, a 5-level residual neural network is designed and trained for each layer from layer 2 to layer 9. Each level of the network has 128 nodes, and the last level is the output level, which is of softMax type. S402, by collecting the status and feature data of vehicles queuing at the intersection, a data set is established to train eight hierarchical neural networks; S403, predicting the minimum speed range and corresponding probability of the target vehicle in the process of reaching the intersection based on 8 neural networks.

10. The method for predicting the minimum speed interval of vehicles at an intersection according to claim 9, characterized in that: In step S402, the neural network corresponding to the 9th layer model is fully trained during the training process, and the neural network corresponding to the 8th to 2nd layer models only trains the last level of the neural network, and the other levels of neural networks reuse the neural network corresponding to the 9th layer model.

Images