Disclosure of Invention
In order to solve the technical problems in the background art, the invention provides the intelligent network vehicle-connected track optimization system and the intelligent network vehicle-connected track optimization method for the shunting area, which can optimize the running track of the intelligent network vehicle-connected in the shunting area, share the lane pressure and improve the traffic efficiency.
In order to achieve the above purpose, the present invention adopts the following technical scheme:
a first aspect of the present invention provides a split area intelligent network train track optimization system, including:
the data perception module is used for acquiring road 3D point cloud data, vehicle size, vehicle position and vehicle routing information;
The system comprises a data perception module, an index calculation module, a server, a data processing module and a data processing module, wherein the data perception module is used for obtaining a shunting area segment where a vehicle is located, a shunting area segment length and a road length occupied by the vehicle, dividing the vehicle type and uploading the vehicle type to the server;
And the judgment decision module is used for determining that the vehicle changes the road from the high-cost segment to the low-cost segment based on the set priority by comparing the cost sum of the adjacent segments and taking the minimum vehicle change times and the minimum traffic stability as targets according to the cost sum of the segments of the shunting area and the cost weights of the segments obtained by the index calculation module.
The data perception module comprises a positioning sensor, an environment perception sensor, a server and a base station, wherein the positioning sensor and the environment perception sensor are used for acquiring road 3D point cloud data, vehicle size, vehicle position and vehicle route information, and the road 3D point cloud data, the vehicle size, the vehicle position and the vehicle route information are uploaded to the server through a base station access network.
The positioning sensor acquires vehicle position and route information, and the environment sensing sensor acquires road 3D point cloud data and vehicle size information.
The split area is a single road area which is equally divided at a set interval perpendicular to a lane line and is equally divided along the lane line, vehicles which are to drive into the exit ramp in the split area are split vehicles, and vehicles which are to drive straight along a main road in the split area are straight vehicles.
The congestion cost is the ratio of the road length occupied by the vehicles of the split area segment to the segment length.
The path cost is the ratio of the number of split vehicles to the total number of vehicles for the split zone segment.
The safety cost is the ratio of the number of large vehicles in the diversion area section to the total number of vehicles, and the large vehicles are vehicles with the length of the vehicle body exceeding a set value in the vehicle size information.
The sum of costs is a weighted sum of congestion costs, path costs, and security costs within the split area segment.
The method comprises the steps of comparing each cost weight of adjacent segments, preferentially controlling corresponding vehicle lane change with higher cost weight sequence in high-cost segments to low-cost segments, and determining the lane change sequence of the vehicles;
And determining the number of lane-changing vehicles required by each vehicle type through the number relation between the sum of the cost and the number of vehicles of each vehicle type.
The high-cost segment is a segment with higher cost sum, and the other segment with lower cost sum is a low-cost segment by calculating the absolute value of the cost sum difference between the segment and the adjacent left segment and the absolute value of the cost sum difference between the segment and the adjacent right segment.
A second aspect of the present invention provides an optimization method based on the above system, comprising the steps of:
The method comprises the steps of setting a range from the upstream of an exit ramp to a split nose end, arranging a plurality of base stations at equal intervals, connecting the base stations with a server, receiving data acquired by the base stations, acquiring vehicle position and vehicle route information by the base stations through a positioning sensor, acquiring road 3D point cloud coordinates and vehicle size by the base stations through an environment sensing sensor, and uploading the road 3D point cloud coordinates, the vehicle size, the vehicle position and the vehicle route information to the server through a base station access network;
Extracting 3D point cloud coordinates of the lane lines, 3D point cloud coordinates of the cross section at a set distance from the exit ramp and the upstream of the exit ramp on the basis of the server, equally dividing the set distance from the upstream of the exit ramp to a road area at the nose end of the diversion area, dividing the set distance from the upstream of the exit ramp to the lane lines at equal intervals perpendicular to the lane lines, and dividing the lane lines along the lane lines to obtain diversion area fragments;
Determining the influence of different shunting area segment congestion costs, path costs and safety costs on driving environment based on a set scoring rule, and obtaining the ratio of the influence sizes of the shunting area segment congestion costs, the path costs and the safety costs, namely the cost sum;
And according to the obtained cost sum of the fragments of the diversion area and the cost weights of the fragments, determining that the vehicle changes lanes from the high-cost fragment to the low-cost fragment based on the set priority by comparing the cost sum of the adjacent fragments and taking the minimum vehicle change times and the minimum traffic stability as targets.
Compared with the prior art, the above technical scheme has the following beneficial effects:
1. the cost problem in the mathematical field is converted by using the physical parameters of the vehicles passing through the diversion area, the different considerations of traffic jam, driving paths and traffic safety of the vehicles in the driving environment are comprehensively considered, and the position distribution of various intelligent network vehicles in the diversion area is optimized by setting different jam cost weights, path cost weights and safety cost weights for different segments of the diversion area, so that the effect of reducing and balancing the sum of adjacent segment cost is achieved.
2. The weight relation of adjacent shunting area segments is considered, when the position distribution of intelligent network vehicles is optimized, vehicles corresponding to higher cost weights are preferentially controlled to be changed into low-cost segments, and the cost sum of the adjacent segments is reduced and balanced by the minimum vehicle change times and the minimum traffic stability.
3. The number of the lane change vehicles required by the lane change sequence is determined by analyzing the number relation between the total cost of the fragments and the number of the vehicles of each vehicle type, a clear decision step is provided for an intelligent network vehicle-connected manager, and the fine control of optimizing the running track of the intelligent network vehicle in the split area is realized.
Detailed Description
The invention will be further described with reference to the drawings and examples.
It should be noted that the following detailed description is exemplary and is intended to provide further explanation of the invention. 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.
It is noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of exemplary embodiments according to the present invention. As used herein, the singular is also intended to include the plural unless the context clearly indicates otherwise, and furthermore, it is to be understood that the terms "comprises" and/or "comprising" when used in this specification are taken to specify the presence of stated features, steps, operations, devices, components, and/or combinations thereof.
As described in the background art, the prior art aims at controlling and co-driving the speed of the intelligent network vehicle at the urban intersections and the confluence areas, but does not optimize the track of the intelligent network vehicle at the diversion areas, and most of the prior art starts from a microscopic level, aims at optimizing the running benefit of a single intelligent network vehicle, and lacks consideration for optimizing the macroscopic layout of the intelligent network vehicle on each lane of the diversion areas, so that the traffic running efficiency is not ideal.
Therefore, the following embodiment provides a system and a method for optimizing the track of the intelligent network vehicle in the diversion area, which can optimize the running track of the intelligent network vehicle in the diversion area, share the pressure of the lane and improve the traffic efficiency.
Embodiment one:
An intelligent network coupling track optimization system for a shunting area, comprising:
the data perception module is used for acquiring road 3D point cloud data, vehicle size, vehicle position and vehicle routing information;
The system comprises a data perception module, an index calculation module, a server, a data processing module and a data processing module, wherein the data perception module is used for obtaining a shunting area segment where a vehicle is located, a shunting area segment length and a road length occupied by the vehicle, dividing the vehicle type and uploading the vehicle type to the server;
And the judgment decision module is used for determining that the vehicle changes the road from the high-cost segment to the low-cost segment based on the set priority by comparing the cost sum of the adjacent segments and taking the minimum vehicle change times and the minimum traffic stability as targets according to the cost sum of the segments of the shunting area and the cost weights of the segments obtained by the index calculation module.
The method comprises the following steps:
(1) Data perception module
The data perception module comprises a positioning sensor, an environment perception sensor, an MEC server, a base station and other devices and is used for collecting and storing road 3D point cloud data, vehicle size, vehicle position and vehicle routing information in real time. The vehicle firstly perceives the 3D point cloud data, the vehicle size, the vehicle position and the vehicle route information through the positioning sensor and the environment perception sensor, and secondly uploads the perceived information to the MEC server through the base station access network, so that information perception and fusion in a certain road range are realized.
The positioning sensor is a vehicle sensor carrying GPS positioning technology and is used for collecting vehicle position and route information in real time.
The environment sensing sensor is a vehicle sensor carrying a laser radar sensing technology and is used for collecting road 3D point cloud data and vehicle size information in real time.
The MEC server is a server carrying a mobile edge computing technology and provides low-delay computing and content service for intelligent network connection.
The route information refers to a route which the vehicle is to travel, can be directly obtained from a positioning sensor, and is uploaded to the MEC server through a base station access network.
(2) Index calculation module
The index calculation module mainly plays a low-time delay calculation function of the MEC server, calculates a shunting area segment where the vehicle is located, a shunting area segment length and a road length occupied by the vehicle in real time based on vehicle position information, 3D point cloud data and vehicle size information uploaded by the data perception module, divides the vehicle into a shunting vehicle and a straight-going vehicle based on uploaded route information, and divides the vehicle into a large vehicle and a small vehicle based on uploaded vehicle size information. And fusing the uploaded data through the MEC server, determining the weight of each cost of the shunting area segment according to an expert scoring method and a sum method, and calculating the sum of the congestion cost, the path cost, the safety cost and the cost of the shunting area segment.
The split area segment refers to a single road area obtained by equally dividing the whole road area from 2000m upstream of the exit ramp to the nose end of the split area at intervals of 500m perpendicular to the lane line and equally dividing the whole road area along the lane line.
The shunting vehicle refers to a vehicle which is about to drive into an exit ramp in a shunting area.
The straight-going vehicle refers to a vehicle which is to be straight-going along a main road in the shunting area.
The congestion cost is the ratio of the road length occupied by the vehicles of the split area segment to the segment length. The calculation method comprises the following steps:
Wherein, Ll is the length of the large-sized vehicle body, nd,l is the number of split large-sized vehicles in the split area segment, ng,l is the number of straight large-sized vehicles in the split area segment, Ls is the length of the small-sized vehicle body, nd,s is the number of split small-sized vehicles in the split area segment, ng,s is the number of straight small-sized vehicles in the split area segment, and L is the length of the split area segment.
The path cost is the ratio of the number of split vehicles to the total number of vehicles in the split zone segment. The calculation method comprises the following steps:
Where n is the total number of vehicles in the split area segment.
The safety cost is the ratio of the number of large vehicles to the total number of vehicles in the split section. The calculation method comprises the following steps:
The cost sum is a weighted sum of congestion cost, path cost and safety cost in the split area segment. The calculation method comprises the following steps:
Zt＝α1Z1+α2Z2+α3Z3 (4)
where α1 is a congestion cost weight, α2 is a path cost weight, and α3 is a security cost weight.
(3) Judgment decision module
The judging and deciding module controls the vehicle to carry out lane change judging and deciding according to the data downloaded by the index calculating module. And the MEC server transmits the fragment cost sum and the fragment cost weights of the shunting area obtained by the index calculation module to the judgment decision module to assist the vehicle to change the road according to the corresponding rules. The method comprises the steps of determining that a vehicle changes from a high-cost segment to a low-cost segment by comparing the cost sum of adjacent segments, determining the vehicle changing sequence by preferentially controlling the vehicle changing corresponding to the higher cost weight in the high-cost segment to the low-cost segment by comparing the cost weights of the adjacent segments, and determining the number of the vehicles required by each vehicle type by analyzing the number relation between the cost sum and the number of the vehicles of each vehicle type. Through the rules, the judgment decision module preferentially controls the vehicle lane change corresponding to the higher cost weight in the high-cost section to the low-cost section, so as to aim at reducing and balancing the cost sum of the adjacent sections through the minimum vehicle lane change times and the minimum traffic stability.
The high-cost segment is characterized in that two segments with larger cost sum difference absolute values are selected by calculating the cost sum difference absolute value of the segment and the adjacent left segment and the cost sum difference absolute value of the segment and the adjacent right segment, wherein the cost sum is high.
The low-cost segments are segments with larger cost sum difference value by calculating the cost sum difference value absolute value of the segment and the adjacent left segment and the cost sum difference value absolute value of the segment and the adjacent right segment, wherein the segments with lower cost sum are selected.
The system comprehensively considers different considerations of traffic jam, driving paths and traffic safety in the driving environment, and optimizes the position distribution of various intelligent network vehicles in the shunting area by setting different jam cost weights, path cost weights and safety cost weights for different segments in the shunting area, thereby achieving the effect of reducing and balancing the sum of adjacent segment costs.
The system considers the weight relation of the adjacent shunting area segments, and when optimizing the position distribution of the intelligent network connected vehicles, the system preferentially controls the vehicles corresponding to the higher cost weights to be changed into low-cost segments, so that the reduction of the minimum vehicle change times and the minimum traffic stability is realized, and the cost sum of the adjacent segments is reduced and balanced.
The system determines the number of the lane change vehicles required by the lane change sequence by analyzing the number relation between the total cost of the fragments and the number of the vehicles of each vehicle type, provides a clear decision step for an intelligent network vehicle connection oriented manager, and realizes the fine control of optimizing the running track of the intelligent network vehicle in the split area.
Embodiment two:
the optimization method based on the system comprises the following steps:
The method comprises the steps of setting a range from the upstream of an exit ramp to a split nose end, arranging a plurality of base stations at equal intervals, connecting the base stations with a server, receiving data acquired by the base stations, acquiring vehicle position and vehicle route information by the base stations through a positioning sensor, acquiring road 3D point cloud coordinates and vehicle size by the base stations through an environment sensing sensor, and uploading the road 3D point cloud coordinates, the vehicle size, the vehicle position and the vehicle route information to the server through a base station access network;
Extracting 3D point cloud coordinates of the lane lines, 3D point cloud coordinates of the cross section at a set distance from the exit ramp and the upstream of the exit ramp on the basis of the server, equally dividing the set distance from the upstream of the exit ramp to a road area at the nose end of the diversion area, dividing the set distance from the upstream of the exit ramp to the lane lines at equal intervals perpendicular to the lane lines, and dividing the lane lines along the lane lines to obtain diversion area fragments;
Determining the influence of different shunting area segment congestion costs, path costs and safety costs on driving environment based on a set scoring rule, and obtaining the ratio of the influence sizes of the shunting area segment congestion costs, the path costs and the safety costs, namely the cost sum;
And according to the obtained cost sum of the fragments of the diversion area and the cost weights of the fragments, determining that the vehicle changes lanes from the high-cost fragment to the low-cost fragment based on the set priority by comparing the cost sum of the adjacent fragments and taking the minimum vehicle change times and the minimum traffic stability as targets.
Specific:
The first step is that the data perception module is required to set 4 base stations at intervals of 500m in the range from 2000m to 2000m on the exit ramp to the road at the split nose, and respectively collect real-time data of vehicles from 2000m to 1500m on the exit ramp, from 1500m to 1000m on the exit ramp, from 1000m to 500m on the exit ramp and from 500m on the exit ramp to the split nose. Meanwhile, an MEC server is required to be configured from the upstream of the exit ramp to the middle section of the split nose end and used for collecting data information uploaded by the fusion base station. The split area roadside equipment arrangement is shown in fig. 1.
The data sensing module is used for storing and updating the vehicle position (xv,yv,zv) and the vehicle routing information in real time in a time-sharing manner through the positioning sensor, and storing and updating the road 3D point cloud coordinates (x, y, z) and the vehicle size (length, width) in a time-sharing manner through the environment sensing sensor. Meanwhile, the network is accessed through the base station, and road 3D point cloud coordinates (x, y, z), vehicle dimensions (length, width), vehicle positions (xv,yv,zv) and vehicle routing information are uploaded to the MEC server.
The index calculation module extracts a 3D point cloud coordinate (xl,yl,zl) of a lane line and 3D point cloud coordinates (xh,yh,zh) of the exit ramp and cross sections of 500m, 1000m, 1500m and 2000m of the upstream of the exit ramp, equally divides the whole road area from 2000m of the upstream of the exit ramp to the nose end of the split area at intervals of 500m, is perpendicular to the lane line, and divides the split area along the lane line to obtain split area segments. The split area segments are shown in fig. 2.
And step four, determining the influence of different shunting area segment congestion costs Z1, path costs Z2 and safety costs Z3 on the driving environment by an expert scoring method, and obtaining the ratio of the influence of shunting area segment congestion costs Z1, path costs Z2 and safety costs Z3, wherein the ratio is shown in table 1.
TABLE 1 ratio of cost effects for the split area segments
| Fragment number of the split area | Congestion cost Z1 | Path cost Z2 | Safety cost Z3 |
| Congestion cost Z1 | 1 | a12 | a13 |
| Path cost Z2 | — | 1 | a23 |
| Safety cost Z3 | — | — | 1 |
Where aij denotes an importance scale of the influence of Zi on the driving environment with respect to the latter, compared to Zj. The importance scale is filled out by the expert with reference to table 2.
Table 2 importance scale evaluation
And fifthly, calculating the congestion cost weight alpha1, the path cost weight alpha2 and the safety cost weight alpha3 of the shunting area segment by using a sum method, wherein the method is shown in a formula (5). After the calculation is completed, the congestion cost weight alpha1, the path cost weight alpha2 and the safety cost weight alpha3 of each split area segment are input into the MEC server.
And step six, an index calculation module calculates corresponding indexes based on the data uploaded by the data perception module by the MEC server. Firstly, the MEC server compares the 3D point cloud coordinates (x, y, z) of the shunting area segment with the vehicle position (xv,yv,zv) to judge the shunting area segment to which the vehicle belongs. When the transverse coordinates x of the 3D points of the split area segments are all greater than or less than the transverse coordinates xv of the vehicle and the longitudinal coordinates y of the 3D points of the split area segments are all greater than or less than the longitudinal coordinates yv of the vehicle, the vehicle does not belong to the split area segments, otherwise, the vehicle belongs to the split area segments. And secondly, multiplying the length (length) of the vehicles in the shunting area segment by the corresponding number of the vehicles and adding the multiplied length to obtain the road length occupied by the vehicles in the shunting area segment by the MEC server, and calculating the congestion cost of the shunting area segment according to the formula (1). And thirdly, dividing the vehicle type by the MEC server according to the vehicle length (length), dividing the vehicle with the vehicle length (length) more than or equal to 6m into large vehicles and small vehicles, and calculating the safety cost of the shunting area segment according to the formula (3). In addition, the MEC server divides the vehicles to be driven into the exit ramp into split vehicles according to the vehicle routing information, divides the vehicles to be driven straight along the main road into straight vehicles, and calculates the split zone path cost according to formula (2). And finally, multiplying the congestion cost, the path cost and the safety cost of the shunting area segments by corresponding weights by the MEC server, and accumulating to obtain the cost sum of the segments. And the MEC server stores and updates the data information obtained by the index calculation module in real time.
And seventhly, the MEC server downloads the congestion cost, the path cost, the safety cost, the cost sum, the congestion cost weight, the path cost weight and the safety cost weight of the fragments obtained by the index calculation module to the judgment decision module through the base station. The decision making module workflow is shown in fig. 3. Firstly, the judgment decision module compares the sum of the costs of adjacent segments positioned on the same cross section of the road, and the cost difference value between the adjacent segments and the current segment is obtained by making a difference. The judging and deciding module selects the adjacent segment with higher absolute value of the difference value of the cost and sets the vehicle to change from the segment with higher cost sum (segment H) to the segment with lower cost sum (segment L). Next, the decision module compares the congestion cost weight α1,H, the path cost weight α2,H, the safety cost weight α3,H of the segment H with the congestion cost weight α1,L, the path cost weight α2,L, and the safety cost weight α3,L of the segment L to determine the lane changing sequence of the vehicle, as shown in table 3.
TABLE 3 lane change sequence for vehicle
And according to the obtained cost sum of the adjacent segments, the lane changing sequence of the vehicles and the cost weights, the judgment decision module calculates the number of lane changing vehicles required by the corresponding lane changing sequence according to the figure 4. In fig. 4, i is the track changing order of the vehicle, Zit,H is the sum of the costs of the track changing pre-track changing segments H of the vehicle with track changing order of i+1, Zit,L is the sum of the costs of the track changing pre-track changing segments L of the vehicle with track changing order of i+1, Δzi is the difference of the costs of the track changing pre-track changing segments H and the track changing pre-track changing segments L of the vehicle with track changing order of i+1, ni+1 is the number of vehicles with track changing order of i+1 required in the track changing segment H, and ni+1,e is the number of vehicles with track changing order of i+1 actually owned in the track changing segment H.
Split cart correspondence equation:
Wherein, α1,H is the congestion cost weight of segment H, nd,l,H is the number of in-segment flow-splitting vehicles in segment H, nd,s,H is the number of in-segment straight vehicles in segment H, nt,H is the total number of in-segment vehicles, ng,l,H is the number of in-segment straight vehicles in segment H, α1,L is the congestion cost weight of segment L, nd,l,L is the number of in-segment flow-splitting vehicles, nd,s,L is the number of in-segment straight vehicles, nt,L is the total number of in-segment vehicles, and ng,l,L is the number of in-segment straight vehicles.
The corresponding equation of the straight-going large-sized vehicle is as follows:
shunting trolley corresponds to the equation:
Where nd,s,H is the number of shunt dollies in segment H and nd,s,L is the number of shunt dollies in segment L.
The corresponding equation of the straight-going small-sized vehicle is as follows:
The judgment decision module distributes the lane change sequence information of the vehicles and the lane change number information required by the corresponding lane change sequence to the vehicles, and sequentially selects the vehicles from front to back by means of an intelligent networking technology, and cooperatively changes lanes to adjacent segments with low cost sum.
The above embodiment comprehensively considers different considerations of traffic jam, driving path and traffic safety in the driving environment, and optimizes the position distribution of various intelligent network vehicles in the shunting area by setting different jam cost weights, path cost weights and safety cost weights for different segments in the shunting area, thereby achieving the effect of reducing and balancing the sum of adjacent segment costs.
The weight relation of adjacent shunting area segments is considered, when the position distribution of intelligent network vehicles is optimized, vehicles corresponding to higher cost weights are preferentially controlled to be changed into low-cost segments, and the cost sum of the adjacent segments is reduced and balanced by the minimum vehicle change times and the minimum traffic stability.
The number of the lane change vehicles required by the lane change sequence is determined by analyzing the number relation between the total cost of the fragments and the number of the vehicles of each vehicle type, a clear decision step is provided for an intelligent network vehicle-connected manager, and the fine control of optimizing the running track of the intelligent network vehicle in the split area is realized.
The above description is only of the preferred embodiments of the present invention and is not intended to limit the present invention, but various modifications and variations can be made to the present invention by those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.