A system and method for optimizing the trajectory of intelligent connected vehicles in a diversion area

Abstract

The present invention relates to a system and method for optimizing the trajectory of intelligent networked vehicles in a diversion area, comprising a data perception module for acquiring road 3D point cloud data, vehicle size, vehicle position and vehicle routing information; an index calculation module for obtaining the diversion area segment where the vehicle is located, the diversion area segment length and the road length occupied by the vehicle according to the information obtained by the data perception module, and classifying the vehicle categories and uploading them to a server; the server determines the weights of each cost of the diversion area segment based on a set scoring rule according to the received information, and calculates the congestion cost, path cost, safety cost and total cost of the diversion area segment; a judgment and decision module for determining that the vehicle changes lanes from a high-cost segment to a low-cost segment based on a set priority according to the total cost of the diversion area segment and the weights of each cost of the segment obtained by the index calculation module, by comparing the total cost of adjacent segments, with the least number of vehicle lane changes and the smallest traffic stability reduction as the goal.

Description

Technical Field

[0001] The present invention relates to the field of intelligent vehicle transportation technology, and in particular to a system and method for optimizing the trajectory of intelligent networked vehicles in a diversion area. Background Art

[0002] The statements in this section merely provide background information related to the present invention and do not necessarily constitute prior art.

[0003] The diversion area is an important node connecting the main line of the highway and the exit ramp. It provides services for diverted vehicles to leave the main line and is one of the places where traffic congestion is prone to occur on highways. In order to successfully leave the main line, diverted vehicles need to adjust their speed to find a gap where they can change lanes and enter the outer lanes, which leads to traffic disorder on the main road and huge driving pressure on the outer lanes. At the same time, under the implementation of the multi-lane highway management strategy, some special vehicles are restricted to driving in the outer lanes, such as trucks, which will further increase the driving pressure on the outer lanes and make it more difficult for diverted vehicles to change lanes.

[0004] At present, intelligent network technology has been gradually introduced, and the distance between vehicles is shorter, the driving speed is faster, and the driving behavior is more consistent. However, the changes in vehicle driving behavior caused by intelligent network technology tend to increase the difficulty of vehicle lane change. First, the shorter distance between vehicles makes it difficult for vehicles to find enough gaps to perform lane changes; second, the faster driving speed of vehicles requires a larger gap for vehicles to change lanes, resulting in more stringent conditions for lane changes; finally, the more consistent driving behavior of vehicles causes multiple shorter headways to appear in succession, making it more difficult for vehicles to complete lane changes in a short period of time. Therefore, the combination of lane management strategy and intelligent network technology in the diversion area will increase the difficulty of vehicle lane changes and reduce traffic operation efficiency.

[0005] Existing technologies focus on speed control and collaborative driving of intelligent connected vehicles at urban intersections and merging areas, but do not optimize the trajectory of intelligent connected vehicles in diverging areas. Moreover, most of the existing technologies start from the micro level, aiming at the optimal driving efficiency of a single intelligent connected vehicle, but lack consideration for the macro layout optimization of intelligent connected vehicles on each lane in the diverging area, resulting in unsatisfactory traffic operation efficiency. Summary of the invention

[0006] In order to solve the technical problems existing in the above-mentioned background technology, the present invention provides a system and method for optimizing the trajectory of intelligent connected vehicles in the diversion area, which can optimize the driving trajectory of intelligent connected vehicles in the diversion area, share lane pressure, and improve traffic efficiency.

[0007] In order to achieve the above object, the present invention adopts the following technical solution:

[0008] A first aspect of the present invention provides a system for optimizing the trajectory of intelligent connected vehicles in a diversion area, comprising:

[0009] Data perception module, used to obtain road 3D point cloud data, vehicle size, vehicle location and vehicle routing information;

[0010] The indicator calculation module obtains the segment of the diversion area where the vehicle is located, the length of the diversion area segment and the length of the road occupied by the vehicle based on the information obtained by the data perception module, and classifies the vehicle into categories and uploads them to the server; the server determines the weight of each cost of the diversion area segment based on the set scoring rules according to the received information, and calculates the congestion cost, path cost, safety cost and total cost of the diversion area segment;

[0011] The judgment and decision-making module, based on the total cost of the diversion area segments and the cost weights of each segment obtained by the indicator calculation module, compares the total cost of adjacent segments, takes the minimum number of vehicle lane changes and the minimum traffic stability reduction as the goal, and determines that the vehicle changes lanes from the high-cost segment to the low-cost segment based on the set priority.

[0012] The data perception module includes positioning sensors, environmental perception sensors, servers and base stations; the road 3D point cloud data, vehicle size, vehicle location and vehicle routing information are obtained through the positioning sensors and environmental perception sensors, and uploaded to the server through the base station access network.

[0013] The positioning sensor obtains the vehicle position and routing information, and the environmental perception sensor obtains the road 3D point cloud data and vehicle size information.

[0014] The diversion area segment is a single road area obtained by dividing the road area from a set distance upstream of the exit ramp to the nose of the diversion area, which is equally divided perpendicular to the lane lines at set intervals and equally divided along the lane lines; vehicles in the diversion area that intend to enter the exit ramp are diversion vehicles; vehicles in the diversion area that intend to go straight along the main road are through vehicles.

[0015] The congestion cost is the ratio of the road length occupied by vehicles in the divergence segment to the segment length.

[0016] The path cost is the ratio of the number of diverted vehicles in the diversion area segment to the total number of vehicles.

[0017] The safety cost is the ratio of the number of large vehicles in the diversion area segment to the total number of vehicles. Large vehicles are vehicles whose body length exceeds the set value in the vehicle size information.

[0018] The total cost is the weighted sum of congestion cost, path cost and safety cost within the diversion area segment.

[0019] By comparing the cost weights of adjacent segments, the corresponding vehicles with higher cost weight rankings in the high-cost segment are preferentially controlled to change lanes to the low-cost segment, and the vehicle lane-changing order is determined;

[0020] The number of lane-changing vehicles required for each model is determined through the quantitative relationship between the total cost and the number of vehicles of each model.

[0021] The high-cost segment is the segment with the higher total cost among the two segments with the larger absolute value of the difference in total cost by calculating the absolute value of the difference in total cost between the segment and the adjacent left segment, and the absolute value of the difference in total cost between the segment and the adjacent right segment; the other segment with the lower total cost is the low-cost segment.

[0022] A second aspect of the present invention provides an optimization method based on the above system, comprising the following steps:

[0023] Multiple base stations are arranged at equal intervals within the set range from the upstream of the exit ramp to the diversion nose. The base stations are connected to the server to receive data obtained by the base stations. The base stations obtain vehicle location and vehicle routing information through positioning sensors. The base stations obtain road 3D point cloud coordinates and vehicle size through environmental perception sensors. The road 3D point cloud coordinates, vehicle size, vehicle location and vehicle routing information are uploaded to the server through the base station access network.

[0024] Based on the 3D point cloud coordinates of the lane line and the 3D point cloud coordinates of the cross section at the set distance upstream of the exit ramp extracted by the server, the road area from the set distance upstream of the exit ramp to the nose of the diverging area is equally divided perpendicular to the lane line at a set interval, and the diverging area segments are obtained by dividing along the lane line;

[0025] Based on the set scoring rules, determine the impact of congestion costs, path costs and safety costs of different diversion area segments on the driving environment, and obtain the ratio of the impact size between the congestion cost, path cost and safety cost of the diversion area segment, which is the total cost; based on the total cost, obtain the congestion cost weight, path cost weight and safety cost weight of the diversion area segment;

[0026] According to the obtained total cost of the diversion area segments and the cost weight of each segment, by comparing the total cost of adjacent segments, with the goal of minimizing the number of vehicle lane changes and the minimum traffic stability reduction, it is determined that the vehicle changes lanes from the high-cost segment to the low-cost segment based on the set priority.

[0027] Compared with the prior art, one or more of the above technical solutions have the following beneficial effects:

[0028] 1. The physical parameters of vehicles passing through the diversion area are converted into cost problems in the mathematical field. The different considerations of vehicles on traffic congestion, driving paths, and traffic safety in the driving environment are comprehensively considered. By setting differentiated congestion cost weights, path cost weights, and safety cost weights for different segments of the diversion area, the position distribution of various types of intelligent connected vehicles in the diversion area is optimized to achieve the effect of reducing and balancing the total cost of adjacent segments.

[0029] 2. The weight relationship between adjacent diversion area segments is taken into consideration. When optimizing the location distribution of intelligent connected vehicles, vehicles with higher cost weights are preferentially controlled to move to low-cost segments, so as to achieve the minimum number of vehicle lane changes and the minimum traffic stability reduction, thereby reducing and balancing the total cost of adjacent segments.

[0030] 3. By analyzing the quantitative relationship between the sum of the segment costs and the number of vehicles of each type, the number of lane-changing vehicles required for the corresponding lane-changing sequence is determined, which provides clear decision-making steps for managers of intelligent connected vehicles and realizes fine control of the optimization of the operating trajectory of intelligent connected vehicles in the diversion area. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] The accompanying drawings in the specification, which constitute a part of the present invention, are used to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute improper limitations on the present invention.

[0032] Figure 1 is a schematic diagram of the arrangement of roadside equipment in a diversion area provided by one or more embodiments of the present invention;

[0033] Figure 2 is a schematic diagram of a flow diversion region fragment provided by one or more embodiments of the present invention;

[0034] Figure 3 is a workflow diagram of a judgment and decision-making module provided by one or more embodiments of the present invention;

[0035] Figure 4 It is a flow chart of outputting the number of lane-changing vehicles provided by one or more embodiments of the present invention. DETAILED DESCRIPTION

[0036] The present invention will be further described below in conjunction with the accompanying drawings and embodiments.

[0037] It should be noted that the following detailed descriptions are exemplary and are intended to provide further explanation of the present invention. Unless otherwise specified, all technical and scientific terms used herein have the same meanings as those commonly understood by those skilled in the art to which the present invention belongs.

[0038] It should be noted that the terms used herein are only for describing specific embodiments and are not intended to limit exemplary embodiments according to the present invention. As used herein, unless the context clearly indicates otherwise, the singular form is also intended to include the plural form. In addition, it should be understood that when the terms "comprising" and / or "including" are used in this specification, it indicates the presence of features, steps, operations, devices, components and / or combinations thereof.

[0039] As described in the background technology, the existing technology focuses on speed control and cooperative driving of intelligent connected vehicles at urban intersections and merging areas, but does not optimize the trajectory of intelligent connected vehicles in diverging areas; and most of the existing technologies start from the micro level, aiming at the optimal driving efficiency of a single intelligent connected vehicle, but lack consideration for the macro layout optimization of intelligent connected vehicles on each lane in the diverging area, resulting in unsatisfactory traffic operation efficiency.

[0040] Therefore, the following embodiments provide a system and method for optimizing the trajectory of intelligent connected vehicles in a diversion area, which can optimize the driving trajectory of intelligent connected vehicles in a diversion area, share lane pressure, and improve traffic efficiency.

[0041] Embodiment 1:

[0042] A system for optimizing the trajectory of intelligent connected vehicles in a diversion area, comprising:

[0043] Data perception module, used to obtain road 3D point cloud data, vehicle size, vehicle location and vehicle routing information;

[0044] The indicator calculation module obtains the segment of the diversion area where the vehicle is located, the length of the diversion area segment and the length of the road occupied by the vehicle based on the information obtained by the data perception module, and classifies the vehicle into categories and uploads them to the server; the server determines the weight of each cost of the diversion area segment based on the set scoring rules according to the received information, and calculates the congestion cost, path cost, safety cost and total cost of the diversion area segment;

[0045] The judgment and decision-making module, based on the total cost of the diversion area segments and the cost weights of each segment obtained by the indicator calculation module, compares the total cost of adjacent segments, takes the minimum number of vehicle lane changes and the minimum traffic stability reduction as the goal, and determines that the vehicle changes lanes from the high-cost segment to the low-cost segment based on the set priority.

[0046] Specifically:

[0047] (1) Data perception module

[0048] The data perception module includes positioning sensors, environmental perception sensors, MEC servers, base stations and other equipment, which are used to collect and store road 3D point cloud data, vehicle size, vehicle location, and vehicle routing information in real time. The vehicle first perceives 3D point cloud data, vehicle size, vehicle location, and vehicle routing information through positioning sensors and environmental perception sensors, and then uploads the perception information to the MEC server through the base station access network to achieve information perception and fusion within a certain road range.

[0049] Positioning sensors refer to vehicle sensors equipped with GPS positioning technology, which are used to collect vehicle location and routing information in real time.

[0050] Environmental perception sensors refer to vehicle sensors equipped with lidar perception technology, which are used to collect road 3D point cloud data and vehicle size information in real time.

[0051] MEC server refers to a server equipped with mobile edge computing technology, which provides low-latency computing and content services for intelligent connected vehicles.

[0052] Routing information refers to the route that the vehicle intends to travel, which can be directly obtained from the positioning sensor and uploaded to the MEC server through the base station access network.

[0053] (2) Index calculation module

[0054] The indicator calculation module mainly uses the low-latency calculation function of the MEC server. Based on the vehicle location information, 3D point cloud data, and vehicle size information uploaded by the data perception module, it calculates the diversion area segment where the vehicle is located, the length of the diversion area segment, and the length of the road occupied by the vehicle in real time; based on the uploaded routing information, the vehicle is divided into diversion vehicles and straight vehicles; based on the uploaded vehicle size information, the vehicle is divided into large vehicles and small vehicles. The uploaded data is integrated through the MEC server, and the weights of each cost of the diversion area segment are determined according to the expert scoring method and the sum method, and the congestion cost, path cost, safety cost, and total cost of the diversion area segment are calculated.

[0055] The diverging area segment refers to a single road area obtained by dividing the entire road area from 2000m upstream of the exit ramp to the nose of the diverging area into equal parts perpendicular to the lane lines at intervals of 500m and along the lane lines.

[0056] Diverted vehicles refer to vehicles in the diverted area that intend to enter the exit ramp.

[0057] Through-going vehicles refer to vehicles that intend to go straight along the main road within the diversion area.

[0058] Congestion cost refers to the ratio of the road length occupied by vehicles in the diversion area segment to the segment length. The calculation method is as follows:

[0059]

[0060] Among them, L l is the length of the large vehicle body, n d,l is the number of large vehicles diverted within the diversion area segment, n g,l is the number of large vehicles going straight in the diversion area segment, L s is the body length of a small car, n d,s is the number of diverted small vehicles in the diverting area segment, n g,s is the number of small vehicles going straight in the diversion area segment, and L is the length of the diversion area segment.

[0061] The path cost is the ratio of the number of diverted vehicles in the diversion area segment to the total number of vehicles. The calculation method is as follows:

[0062]

[0063] Where n is the total number of vehicles in the diversion area segment.

[0064] Safety cost refers to the ratio of the number of large vehicles to the total number of vehicles in the diversion area segment. The calculation method is as follows:

[0065]

[0066] The total cost refers to the weighted sum of congestion cost, path cost and safety cost within the diversion area segment. The calculation method is as follows:

[0067] Z t =α1Z1+α2Z2+α3Z3 (4)

[0068] Among them, α1 is the congestion cost weight, α2 is the path cost weight, and α3 is the safety cost weight.

[0069] (3) Judgment and decision module

[0070] The judgment and decision module controls the vehicle to make lane change decisions based on the data downloaded by the index calculation module. The MEC server transmits the sum of the cost of the diversion area segments and the cost weights of each segment obtained by the index calculation module to the judgment and decision module to assist the vehicle in changing lanes according to the corresponding rules. The judgment and decision module determines that the vehicle changes lanes from the high-cost segment to the low-cost segment by comparing the sum of the costs of adjacent segments; by comparing the cost weights of the adjacent segments, the vehicle corresponding to the higher cost weight in the high-cost segment is preferentially controlled to change lanes to the low-cost segment, thereby determining the order of vehicle lane changes; by analyzing the quantitative relationship between the sum of the costs and the number of vehicles of each model, the number of lane change vehicles required for each model is determined. Through the above rules, the judgment and decision module preferentially controls the vehicle corresponding to the higher cost weight in the high-cost segment to change lanes to the low-cost segment, striving to reduce and balance the sum of the costs of adjacent segments through the minimum number of vehicle lane changes and the minimum traffic stability reduction.

[0071] A high-cost segment refers to a segment with a higher total cost, which is obtained by calculating the absolute value of the difference between the total cost of the segment and the adjacent left segment, and the absolute value of the difference between the total cost of the segment and the adjacent right segment, and selecting two segments with the larger absolute value of the difference between the total cost.

[0072] A low-cost segment refers to a segment with a lower total cost, which is obtained by calculating the absolute value of the difference between the total cost of the segment and the adjacent left segment, and the absolute value of the difference between the total cost of the segment and the adjacent right segment, and selecting two segments with the larger absolute value of the difference in total cost.

[0073] The above system comprehensively considers the different considerations of vehicles on traffic congestion, driving routes and traffic safety in the driving environment. By setting differentiated congestion cost weights, path cost weights and safety cost weights for different segments of the diversion area, it optimizes the position distribution of various types of intelligent connected vehicles in the diversion area, so as to achieve the effect of reducing and balancing the total cost of adjacent segments.

[0074] The above system takes into account the weight relationship between adjacent diversion area segments. When optimizing the location distribution of intelligent connected vehicles, it gives priority to controlling vehicles corresponding to higher cost weights to move to low-cost segments, thereby achieving the minimum number of vehicle lane changes and the minimum traffic stability reduction, reducing and balancing the total cost of adjacent segments.

[0075] The above system determines the number of lane-changing vehicles required for the corresponding lane-changing sequence by analyzing the quantitative relationship between the sum of the segment costs and the number of vehicles of each type. It provides clear decision-making steps for managers of intelligent connected vehicles and realizes fine control over the optimization of the operating trajectory of intelligent connected vehicles in the diversion area.

[0076] Embodiment 2:

[0077] The optimization method based on the above system includes the following steps:

[0078] Multiple base stations are arranged at equal intervals within the set range from the upstream of the exit ramp to the diversion nose. The base stations are connected to the server to receive data obtained by the base stations. The base stations obtain vehicle location and vehicle routing information through positioning sensors. The base stations obtain road 3D point cloud coordinates and vehicle size through environmental perception sensors. The road 3D point cloud coordinates, vehicle size, vehicle location and vehicle routing information are uploaded to the server through the base station access network.

[0079] Based on the 3D point cloud coordinates of the lane line and the 3D point cloud coordinates of the cross section at the set distance upstream of the exit ramp extracted by the server, the road area from the set distance upstream of the exit ramp to the nose of the diverging area is equally divided perpendicular to the lane line at a set interval, and the diverging area segments are obtained by dividing along the lane line;

[0080] Based on the set scoring rules, determine the impact of congestion costs, path costs and safety costs of different diversion area segments on the driving environment, and obtain the ratio of the impact size between the congestion cost, path cost and safety cost of the diversion area segment, which is the total cost; based on the total cost, obtain the congestion cost weight, path cost weight and safety cost weight of the diversion area segment;

[0081] According to the obtained total cost of the diversion area segments and the cost weight of each segment, by comparing the total cost of adjacent segments, with the goal of minimizing the number of vehicle lane changes and the minimum traffic stability reduction, it is determined that the vehicle changes lanes from the high-cost segment to the low-cost segment based on the set priority.

[0082] Specific:

[0083] Step 1: Data perception module, within the road range from 2000m upstream of the exit ramp to the diversion nose, 4 base stations are set up at intervals of 500m to collect real-time data of vehicles from 2000m to 1500m upstream of the exit ramp, 1500m to 1000m upstream of the exit ramp, 1000m to 500m upstream of the exit ramp, and 500m upstream of the exit ramp to the diversion nose. At the same time, an MEC server needs to be configured from the upstream of the exit ramp to the middle of the diversion nose to collect and integrate data information uploaded by the base station. The layout of the roadside equipment in the diversion area is as follows: Figure 1 shown.

[0084] Step 2: Data perception module uses positioning sensors to detect the vehicle position (x v ,y v ,z v ) and vehicle routing information are stored in different time periods and updated in real time; through environmental perception sensors, the road 3D point cloud coordinates (x, y, z) and vehicle dimensions (length, width) are stored in different time periods and updated in real time. At the same time, the base station is connected to the network to store the road 3D point cloud coordinates (x, y, z), vehicle dimensions (length, width), and vehicle position (x v ,y v ,z v ), vehicle routing information is uploaded to the MEC server.

[0085] Step 3: Index calculation module, based on the MEC server, extracts the 3D point cloud coordinates (x l ,y l ,z l ) and the 3D point cloud coordinates of the exit ramp and the cross sections 500m, 1000m, 1500m, and 2000m upstream of the exit ramp (x h ,y h ,z h ), the whole road area from 2000m upstream of the exit ramp to the nose of the diversion area is divided into equal parts perpendicular to the lane line at intervals of 500m, and the diversion area segments are obtained along the lane line. Figure 2 shown.

[0086] Step 4: Through the expert scoring method, clarify the impact of different diversion area segment congestion costs Z1, path costs Z2 and safety costs Z3 on the driving environment, and obtain the impact ratio between the diversion area segment congestion cost Z1, path cost Z2 and safety cost Z3, as shown in Table 1.

[0087] Table 1 Cost impact ratios of various segments in the diversion area

[0088] Sequence number of the split zone segment <![CDATA[Congestion cost Z1]]> <![CDATA[Path cost Z2]]> <![CDATA[Safety cost Z3]]> <![CDATA[Congestion cost Z1]]> 1 <![CDATA[a 12 ]]> <![CDATA[a 13 ]]> <![CDATA[Path cost Z2]]> — 1 <![CDATA[a 23 ]]> <![CDATA[Safety cost Z3]]> — — 1

[0089] Among them, a ij Represents Z i With Z j The importance scale of the former relative to the latter on the driving environment is filled in by the experts with reference to Table 2.

[0090] Table 2 Importance scale evaluation

[0091]

[0092] Step 5: Use the sum method to calculate the congestion cost weight α1, path cost weight α2, and safety cost weight α3 of the diversion area segment, as shown in formula (5). After the calculation is completed, the congestion cost weight α1, path cost weight α2, and safety cost weight α3 of each diversion area segment are input into the MEC server.

[0093]

[0094] Step 6: The indicator calculation module calculates the corresponding indicators based on the data uploaded by the data perception module from the MEC server. First, the MEC server compares the 3D point cloud coordinates (x, y, z) of the split area segment with the vehicle position (x v ,y v ,z v ), determine the segment of the diversion area to which the vehicle belongs. When the lateral coordinates x of the 3D points of the diversion area segment are greater than or less than the lateral coordinate x of the vehicle v The longitudinal coordinates y of the 3D points in the diverging area are all greater than or less than the longitudinal coordinate y of the vehicle. v When , the vehicle does not belong to the diversion area segment, otherwise, the vehicle belongs to the diversion area segment. Secondly, the MEC server multiplies the length of the vehicle in the diversion area segment by the corresponding number of vehicles and adds them up to obtain the road length occupied by the vehicles in the diversion area segment, and calculates the congestion cost of the diversion area segment according to formula (1). Thirdly, the MEC server divides the vehicle types according to the length of the vehicle, and divides the vehicles with a length of ≥6m into large vehicles and the rest into small vehicles, and calculates the safety cost of the diversion area segment according to formula (3). In addition, the MEC server divides the vehicles that intend to enter the exit ramp into diversion vehicles and the vehicles that intend to go straight along the main road into straight vehicles according to the vehicle routing information, and calculates the path cost of the diversion area according to formula (2). Finally, the MEC server multiplies the congestion cost, path cost and safety cost of the diversion area segment by the corresponding weights and adds them up to obtain the total cost of the segment. The MEC server stores and updates the data information obtained by the indicator calculation module in real time.

[0095] Step 7: The MEC server downloads the congestion cost, path cost, safety cost, total cost, congestion cost weight, path cost weight and safety cost weight of the fragment obtained by the indicator calculation module to the judgment and decision-making module through the base station. The judgment and decision-making module workflow is as follows: Figure 3 As shown. First, the judgment and decision module compares the sum of the costs of adjacent segments in the same cross section of the road, and obtains the cost difference between the adjacent segment and the current segment by subtraction. The judgment and decision module selects the adjacent segment with a higher absolute value of the cost difference between the current segment and the segment, and sets the vehicle to change lanes from the segment with a higher total cost (segment H) to the segment with a lower total cost (segment L). Secondly, the judgment and decision module compares the congestion cost weight of segment H. 1,H , path cost weight α 2,H , safety cost weight α 3,H The congestion cost weight of segment L is the congestion cost weight α 1,L , path cost weight α 2,L , safety cost weight α 3,L , determine the lane changing order of the vehicles, as shown in Table 3.

[0096] Table 3 Vehicle lane change sequence

[0097]

[0098]

[0099] According to the sum of the costs of the adjacent segments, the order of vehicle lane changes and the weights of each cost, the decision module determines the Figure 4 The number of lane-changing vehicles required for the corresponding lane-changing sequence is calculated. Figure 4 In the equation, i is the order in which the vehicle changes lanes, and Z i t,H is the sum of the costs of the segment H before the lane change of the vehicle with lane change order i+1, Z i t,L is the sum of the costs of the segment L before the lane change of the vehicle with lane change order i+1, ΔZ i is the cost difference between segment H and segment L before the lane change of the vehicle with lane change order i+1, n i+1 is the number of vehicles that need to change lanes in sequence i+1 in segment H, n i+1,e is the number of vehicles with lane change order i+1 in segment H.

[0100] The corresponding equation for diverting large vehicles is:

[0101]

[0102] Among them, α 1,H is the congestion cost weight of segment H, n d,l,H is the number of large vehicles diverted in segment H, nd,s,H is the number of large vehicles going straight in segment H, n t,H is the total number of vehicles in segment H, n g,l,H is the number of large vehicles going straight in segment H, α 1,L is the congestion cost weight of segment L, n d,l,L is the number of large vehicles diverted in segment L, n d,s,L is the number of large vehicles going straight in segment L, n t,L is the total number of vehicles in segment L, n g,l,L is the number of large vehicles going straight in segment L.

[0103] The corresponding formula for a large straight-moving vehicle is:

[0104]

[0105] The corresponding formula for diverting small cars is:

[0106]

[0107] Among them, n d,s,H is the number of small cars diverted in segment H, n d,s,L is the number of small cars diverted within segment L.

[0108] The corresponding formula for a small car driving straight:

[0109]

[0110] The judgment and decision module distributes the lane change sequence information and the number of lane-changing vehicles required for the corresponding lane change sequence to the vehicles, and uses intelligent network connection technology to select vehicles in sequence from the front to the back, and coordinate lane changes to adjacent segments with lower total costs.

[0111] The above embodiments comprehensively consider the different considerations of vehicles for traffic congestion, driving paths, and traffic safety in the driving environment. By setting differentiated congestion cost weights, path cost weights, and safety cost weights for different segments of the diversion area, the position distribution of various types of intelligent connected vehicles in the diversion area is optimized to achieve the effect of reducing and balancing the total cost of adjacent segments.

[0112] The weight relationship between adjacent diversion area segments is taken into consideration. When optimizing the location distribution of intelligent connected vehicles, vehicles with higher cost weights are preferentially controlled to move to low-cost segments, so as to achieve the minimum number of vehicle lane changes and the minimum traffic stability reduction, thereby reducing and balancing the total cost of adjacent segments.

[0113] By analyzing the quantitative relationship between the sum of the segment costs and the number of vehicles of each type, the number of lane-changing vehicles required for the corresponding lane-changing sequence is determined, which provides clear decision-making steps for managers of intelligent connected vehicles and achieves fine control over the optimization of the operating trajectories of intelligent connected vehicles in the diversion area.

[0114] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A system for optimizing the trajectory of intelligent connected vehicles in a diversion area, characterized by: include: Data perception module, used to obtain road 3D point cloud data, vehicle size, vehicle location and vehicle routing information; The indicator calculation module obtains the segment of the diversion area where the vehicle is located, the length of the diversion area segment and the length of the road occupied by the vehicle based on the information obtained by the data perception module, and classifies the vehicle into categories and uploads them to the server; the server determines the weight of each cost of the diversion area segment based on the set scoring rules according to the received information, and calculates the congestion cost, path cost, safety cost and total cost of the diversion area segment; The judgment and decision module determines that the vehicle changes lanes from the high-cost segment to the low-cost segment based on the set priority, based on the total cost of the segment and the cost weight of each segment obtained by the indicator calculation module, by comparing the total cost of adjacent segments, with the goal of minimizing the number of vehicle lane changes and the minimum traffic stability reduction; By comparing the cost weights of adjacent segments, vehicles with higher cost weights in high-cost segments are prioritized to change lanes to low-cost segments to determine the order of vehicle lane change; the number of lane-changing vehicles required for each model is determined through the quantitative relationship between the total cost and the number of vehicles of each model.

2. The intelligent connected vehicle trajectory optimization system in a diversion area according to claim 1, characterized in that: The data perception module includes a positioning sensor, an environmental perception sensor, a server and a base station; the road 3D point cloud data, vehicle size, vehicle location and vehicle routing information are obtained through the positioning sensor and the environmental perception sensor, and uploaded to the server through the base station access network.

3. The intelligent connected vehicle trajectory optimization system in a diversion area according to claim 1, characterized in that: The diversion area segment is a road area from a set distance upstream of the exit ramp to the nose of the diversion area, which is equally divided perpendicular to the lane line at a set interval and equally divided along the lane line to obtain a single road area; a vehicle in the diversion area that intends to enter the exit ramp is a diversion vehicle; Vehicles that intend to go straight along the main road in the diversion area are straight-going vehicles.

4. The intelligent connected vehicle trajectory optimization system in a diversion area according to claim 1, characterized in that: The congestion cost is the ratio of the road length occupied by vehicles in the diversion area segment to the segment length.

5. The intelligent connected vehicle trajectory optimization system in a diversion area according to claim 1, characterized in that: The path cost is the ratio of the number of diverted vehicles in the diverting area segment to the total number of vehicles.

6. The intelligent connected vehicle trajectory optimization system in a diversion area according to claim 1, characterized in that: The safety cost is the ratio of the number of large vehicles to the total number of vehicles in the diversion area segment, and the large vehicle is a vehicle whose body length exceeds a set value in the vehicle size information.

7. The intelligent connected vehicle trajectory optimization system in a diversion area according to claim 1, characterized in that: The total cost is the weighted sum of the congestion cost, path cost and safety cost within the diversion area segment.

8. The intelligent connected vehicle trajectory optimization system in a diversion area according to claim 1, characterized in that: The high-cost segment is a segment with a higher total cost among the two segments with the larger absolute value of the difference in total cost by calculating the absolute value of the difference in total cost between the segment and the adjacent left segment, and the absolute value of the difference in total cost between the segment and the adjacent right segment; the other segment with a lower total cost is a low-cost segment.

9. A method based on the system according to any one of claims 1 to 8, characterized in that: The following steps are involved: In a set range from the upstream of the exit ramp to the diversion nose, multiple base stations are arranged at equal intervals, and the base stations are connected to the server to receive data acquired by the base stations; The base station obtains the vehicle position and vehicle routing information through the positioning sensor; the base station obtains the road 3D point cloud coordinates and vehicle size through the environmental perception sensor; the base station accesses the network and uploads the road 3D point cloud coordinates, vehicle size, vehicle position and vehicle routing information to the server; Based on the 3D point cloud coordinates of the lane line and the 3D point cloud coordinates of the cross section at the set distance upstream of the exit ramp extracted by the server, the road area from the set distance upstream of the exit ramp to the nose of the diverging area is equally divided perpendicular to the lane line at a set interval, and the diverging area segments are obtained by dividing along the lane line; Based on the set scoring rules, determine the impact of congestion costs, path costs and safety costs of different diversion area segments on the driving environment, obtain the impact ratio between the congestion costs, path costs and safety costs of the diversion area segments, use the sum method to calculate the congestion cost weights, path cost weights and safety cost weights of the diversion area segments, multiply the congestion cost, path cost and safety cost of the diversion area segments by the corresponding weights and add them up to get the total cost of the segments; According to the obtained sum of segment costs in the diversion area and the weight of each segment cost, by comparing the sum of the costs of adjacent segments, with the goal of minimizing the number of vehicle lane changes and minimizing traffic stability reduction, determine that the vehicle changes lanes from the high-cost segment to the low-cost segment based on the set priority; By comparing the cost weights of adjacent segments, vehicles with higher cost weights in high-cost segments are prioritized to change lanes to low-cost segments to determine the order of vehicle lane change; the number of lane-changing vehicles required for each model is determined through the quantitative relationship between the total cost and the number of vehicles of each model.

Images