Lane selection method and device based on automatic driving

Abstract

The application discloses a lane selection method and device based on automatic driving, and the method comprises the following steps: dividing a target road into multiple road sections; acquiring the average vehicle passing speed of each lane of each road section to be passed within a current unit time; determining the average vehicle passing speed of each lane of each road section within a next unit time according to the average vehicle passing speed of each lane of each road section within the current unit time and a linear regression model; calculating the shortest time from the current position to reach and pass the last road section according to a preset lane selection condition, the road section length of each road section, and the average vehicle passing speed of each lane of each road section within the next unit time, and determining the lane sequence corresponding to the shortest time as a target lane sequence required by a vehicle to pass the road sections to be passed. Driving according to the target lane sequence with the shortest passing time can improve the passing efficiency.

Description

Technical Field

[0001] This invention relates to the field of autonomous driving, and more specifically, to a lane selection method and apparatus based on autonomous driving. Background Technology

[0002] Autonomous driving, also known as driverless driving, relies on the collaborative efforts of artificial intelligence, computer vision, radar, monitoring devices, and global positioning systems to enable computers to operate motor vehicles safely and automatically without any human intervention. Route planning is one of the core technologies of autonomous driving, determining the shortest route from the origin to the destination. However, a road often includes multiple lanes, and current route planning technologies only determine where lane changes are necessary based on the destination, without considering comprehensive information such as road conditions. For example, when transitioning from a highway to a regular road, a lane must be used to exit the highway, and to reach the destination, a lane must be used to reach the destination. Lane changes are only planned in situations where a lane change is mandatory, but not during the journey itself, based on road conditions. Furthermore, on long roads, a single lane is rarely consistently clear. If vehicles travel in only one lane, congestion in that lane can lead to low traffic efficiency. Summary of the Invention

[0003] This invention provides a lane selection method and apparatus based on autonomous driving, which can pre-plan the target lane sequence with the shortest travel time, so as to perform autonomous driving based on the target lane sequence, thereby improving traffic efficiency.

[0004] The specific technical solution is as follows:

[0005] In a first aspect, embodiments of the present invention provide a lane selection method based on autonomous driving, the method comprising:

[0006] Divide the target road into multiple road segments;

[0007] Get the average vehicle speed of each lane in each road segment to be traversed in the current unit of time;

[0008] Based on the average vehicle speed of each lane in each road segment in the current unit time and the linear regression model, the average vehicle speed of each lane in each road segment in the next unit time is determined. The linear regression model is a model trained based on the historical average vehicle speed and used to determine the average vehicle speed of each lane in each road segment in the next unit time. The historical average vehicle speed is the average vehicle speed in a historical unit time.

[0009] Based on preset lane selection conditions, the length of each road segment, and the average vehicle speed of each lane in each road segment in the next unit time, the shortest time to reach and pass through the last road segment from the current position is calculated, and the lane sequence corresponding to the shortest time is determined as the target lane sequence required for the vehicle to pass through the road segment to be passed. The preset lane selection conditions are that when moving from the current lane of the current road segment to the next road segment, the selectable lanes include the current lane and the lane adjacent to the current lane.

[0010] Optionally, the target road can be divided into multiple road segments, including:

[0011] Based on the historical average vehicle speed on the target road, the target road is divided into multiple road segments, such that within a preset road segment length range, the historical average vehicle speed is positively correlated with the road segment length.

[0012] Optionally, based on the historical average vehicle speed on the target road, the target road is divided into multiple road segments, such that within a preset road segment length range, the historical average vehicle speed is positively correlated with the road segment length, including:

[0013] When the historical average vehicle speed is less than or equal to the first average vehicle speed threshold, the corresponding road segment length is the first length.

[0014] When the historical average vehicle speed is greater than the first average vehicle speed threshold and less than the second average vehicle speed threshold, the corresponding road segment length is proportional to the historical average vehicle speed.

[0015] When the historical average vehicle speed is greater than the second average vehicle speed threshold, the corresponding road segment length is the second length.

[0016] Wherein, the first average vehicle speed threshold is less than the second average vehicle speed threshold, and the first length is less than the second length.

[0017] Optionally, obtain the average vehicle speed of each lane of each road segment to be traversed in the current unit of time, including:

[0018] The instantaneous vehicle speeds of all vehicles in each lane of each road segment to be traversed, collected by roadside sensing devices, are obtained within the current unit of time.

[0019] For each lane of each road segment to be traversed, the average vehicle speed of each lane in the current unit of time is calculated based on the number of vehicles at each sampling moment in the current unit of time, the number of samplings in the current unit of time, and the instantaneous vehicle speed.

[0020] Optionally, based on the average vehicle speed of each lane in each road segment in the current unit of time and a linear regression model, the average vehicle speed of each lane in each road segment in the next unit of time is determined, including:

[0021] The average vehicle speed V of the i-th lane in the k-th road segment during the (n+1)-th unit time is calculated based on the linear regression model. i，k (n+1);

[0022] The linear regression model is as follows:

[0023] V i，k (n+1)=u+[a1 a2 a3 a4 a5 a6 a7 a8 a9]×[V i-1，k-1 (n)V i-1，k (n)V i-1，k+1 (n)V i，k-1 (n)V i，k (n)V i，k+1 (n)V i+1，k-1 (n)V i+1，k (n)V i+1xk+1 (n)] T

[0024] Where u, a1, a2, a3, a4, a5, a6, a7, a8, and a9 are the parameters of the linear regression model, and V i-1，k-1 (n) represents the average vehicle speed of the (i-1)th lane of the (k-1)th road segment in the nth unit of time, V i-1，k (n) represents the average vehicle speed of the (i-1)th lane of the k-th road segment in the nth unit of time. i-1，k+1 (n) represents the average vehicle speed of the (i-1)th lane in the (k+1)th road segment during the nth unit of time, V i，k-1 (n) represents the average vehicle speed of the i-th lane in the (k-1)-th road segment during the n-th unit of time, V i，k (n) represents the average vehicle speed of the i-th lane in the k-th road segment during the n-th unit of time, V i，k+1 (n) represents the average vehicle speed of the i-th lane in the (k+1)-th road segment during the n-th unit of time, V i+1，k-1(n) represents the average vehicle speed of the (i+1)th lane of the (k-1)th road segment in the nth unit of time, V i+1，k (n) represents the average vehicle speed of the (i+1)th lane of the k-th road segment in the n-th unit of time, V i+1，k+1 (n) represents the average vehicle speed of the (i+1)th lane of the (k+1)th road segment in the nth unit of time.

[0025] Optionally, based on preset lane selection conditions, the length of each road segment, and the average vehicle speed of each lane in each road segment within the next unit of time, the shortest time to reach and pass through the last road segment from the current location is calculated, including:

[0026] Based on the road segment length of each road segment and the average vehicle speed of each lane in each road segment in the next unit of time, calculate the estimated time for a vehicle to pass through each lane of each road segment.

[0027] Based on the preset lane selection conditions and the estimated time to pass through each lane of each road segment, calculate the shortest time to reach and pass through the last road segment from the current location.

[0028] Optionally, based on the preset lane selection conditions and the estimated time to traverse each lane of each road segment, the shortest time to reach and traverse the last road segment from the current location is calculated, including:

[0029] Based on the preset formula and the estimated time for each lane of each road segment, the shortest time to reach and pass through the last road segment from the current position is obtained for each lane. The minimum value among the shortest times for each lane is determined as the shortest time to reach and pass through the last road segment.

[0030] Wherein, the preset formula is

[0031] S i，k =min{S i-1，k-1 S i，k-1 S i+1，k-1}+T i，k ,

[0032] Among them, S i，k S is the shortest time to reach and pass through the i-th lane of the k-th road segment from the current position. i-1，k-1 S is the shortest time to reach and pass through the (i-1)th lane of the (k-1)th road segment from the current position. i，k-1 S is the shortest time to reach and pass through the i-th lane of the (k-1)-th road segment from the current position. i+1，k-1T is the shortest time to reach and pass through the (i+1)th lane of the (k-1)th road segment from the current position. i，k The estimated time to pass through the i-th lane of the k-th road segment.

[0033] Optionally, the method further includes:

[0034] When the target lane of the starting road segment is a fixed lane, the estimated time of the other lanes of the starting road segment is set to a preset estimated time, which is greater than the estimated time of all lanes.

[0035] When the target lane of the last road segment is a fixed lane, the estimated time of the other lanes of the last road segment is set to the preset estimated time.

[0036] Optionally, the method further includes:

[0037] During the driving process based on the target lane sequence, when a traffic anomaly is detected or the lane planning update time is reached, the process returns to obtain the average vehicle speed of each lane of each road segment to be traversed in the current unit of time, so as to obtain the updated target lane sequence, and continues driving based on the updated target lane sequence.

[0038] Secondly, embodiments of the present invention provide a lane selection device based on autonomous driving, the device comprising:

[0039] Dividing units are used to divide a target road into multiple road segments;

[0040] The acquisition unit is used to acquire the average vehicle speed of each lane in each road segment to be traversed in the current unit of time.

[0041] The first determining unit is used to determine the average vehicle speed of each lane of each road segment in the next unit time based on the average vehicle speed of each lane of each road segment in the current unit time and a linear regression model. The linear regression model is a model trained based on historical average vehicle speeds to determine the average vehicle speed of each lane of each road segment in the next unit time. The historical average vehicle speed is the average vehicle speed in historical unit time.

[0042] The calculation unit is used to calculate the shortest time to reach and pass through the last road segment from the current location based on preset lane selection conditions, the road segment length of each road segment, and the average vehicle speed of each lane of each road segment in the next unit of time.

[0043] The second determining unit is used to determine the lane sequence corresponding to the shortest time as the target lane sequence required for the vehicle to pass through the road segment to be passed. The preset lane selection condition is that when moving from the current lane of the current road segment to the next road segment, the selectable lanes include the current lane and the lane adjacent to the current lane.

[0044] Optionally, the division unit is used to divide the target road into multiple road segments based on the historical average vehicle speed on the target road, such that the historical average vehicle speed is positively correlated with the road segment length within a preset road segment length range.

[0045] Optionally, the division unit is used to determine the first length of the corresponding road segment when the historical average vehicle speed is less than or equal to the first average vehicle speed threshold.

[0046] When the historical average vehicle speed is greater than the first average vehicle speed threshold and less than the second average vehicle speed threshold, the corresponding road segment length is proportional to the historical average vehicle speed.

[0047] When the historical average vehicle speed is greater than the second average vehicle speed threshold, the corresponding road segment length is the second length.

[0048] Wherein, the first average vehicle speed threshold is less than the second average vehicle speed threshold, and the first length is less than the second length.

[0049] Optionally, the acquisition unit includes:

[0050] The acquisition module is used to acquire the instantaneous vehicle speeds of all vehicles in each lane of each road segment to be traversed, collected by the roadside sensing device within the current unit of time.

[0051] The first calculation module is used to calculate the average vehicle speed of each lane in the current unit time for each lane of each road segment to be traversed, based on the number of vehicles at each sampling moment in the current unit time, the number of samplings in the current unit time, and the instantaneous vehicle speed.

[0052] Optionally, the first determining unit is configured to calculate, according to the first formula, the average vehicle speed V of the i-th lane of the k-th road segment in the (n+1)-th unit time. i，k (n+1);

[0053] The first formula is

[0054] V i，k(n+1)=u+[a1 a2 a3 a4 a5 a6 a7 a8 a9]×[V i-1，k-1 (n) V i-1，k (n)V i-1，k+1 (n) V i，k-1 (n) V i，k (n) V i，k+1 (n) V i+1，k-1 (n) V i+1，k (n) V i+1，k+1 (n)] T

[0055] Where u, a1, a2, a3, a4, a5, a6, a7, a8, and a9 are the parameters of the linear regression model, and V i-1，k-1 (n) represents the average vehicle speed of the (i-1)th lane of the (k-1)th road segment in the nth unit of time, V i-1，k (n) represents the average vehicle speed of the (i-1)th lane of the k-th road segment in the nth unit of time. i-1，k+1 (n) represents the average vehicle speed of the (i-1)th lane in the (k+1)th road segment during the nth unit of time, V i，k-1 (n) represents the average vehicle speed of the i-th lane in the (k-1)-th road segment during the n-th unit of time, V i，k (n) represents the average vehicle speed of the i-th lane in the k-th road segment during the n-th unit of time, V i，k+1 (n) represents the average vehicle speed of the i-th lane in the (k+1)-th road segment during the n-th unit of time, V i+1，k-1 (n) represents the average vehicle speed of the (i+1)th lane of the (k-1)th road segment in the nth unit of time, V i+1，k (n) represents the average vehicle speed of the (i+1)th lane of the k-th road segment in the n-th unit of time, V i+1，k+1 (n) represents the average vehicle speed of the (i+1)th lane of the (k+1)th road segment in the nth unit of time.

[0056] Optionally, the computing unit includes:

[0057] The second calculation module is used to calculate the estimated time for a vehicle to pass through each lane of each road segment based on the road segment length of each road segment and the average vehicle speed of each lane of each road segment in the next unit of time.

[0058] The third calculation module is used to calculate the shortest time from the current position to the last road segment based on the preset lane selection conditions and the estimated time to pass through each lane of each road segment.

[0059] Optionally, a third calculation module is used to iteratively calculate based on a preset formula and the estimated time for each lane of each road segment to obtain the shortest time from the current position to reach and pass through each lane of the last road segment, and to determine the minimum value among the shortest times of each lane as the shortest time to reach and pass through the last road segment.

[0060] Wherein, the preset formula is

[0061] S i，k =min{S i-1，k-1 S i，k-1 S i+1，k-1}+T i，k ,

[0062] Among them, S i，k S is the shortest time to reach and pass through the i-th lane of the k-th road segment from the current position. i-1，k-1 S is the shortest time to reach and pass through the (i-1)th lane of the (k-1)th road segment from the current position. i，k-1 S is the shortest time to reach and pass through the i-th lane of the (k-1)-th road segment from the current position. i+1，k-1 T is the shortest time to reach and pass through the (i+1)th lane of the (k-1)th road segment from the current position. i，k The estimated time to pass through the i-th lane of the k-th road segment.

[0063] Optionally, the device further includes:

[0064] The setting unit is configured to, when the target lane of the starting road segment is a fixed lane, set the estimated time of the other lanes of the starting road segment to a preset estimated time, wherein the preset estimated time is greater than the estimated time of all lanes; and when the target lane of the last road segment is a fixed lane, set the estimated time of the other lanes of the last road segment to the preset estimated time.

[0065] Optionally, the device further includes:

[0066] The return execution unit is used to, during the driving process based on the target lane sequence, when a traffic anomaly event is detected or the lane planning update time is reached, return to execute the step of obtaining the average vehicle speed of each lane of each road segment to be passed in the current unit time, so as to obtain the updated target lane sequence, and continue driving based on the updated target lane sequence.

[0067] Thirdly, embodiments of the present invention provide a storage medium storing executable instructions thereon, which, when executed by a processor, cause the processor to implement the method described in the first aspect.

[0068] Fourthly, embodiments of the present invention provide an autonomous driving vehicle, comprising:

[0069] One or more processors;

[0070] Storage device for storing one or more programs.

[0071] Wherein, when the one or more programs are executed by the one or more processors, the one or more processors implement the method described in the first aspect.

[0072] As can be seen from the above, the lane selection method and device based on autonomous driving provided in this embodiment of the invention can first divide the target road into multiple road segments, then obtain the average vehicle speed of each lane in each road segment to be traversed in the current unit time, and predict the average vehicle speed of each lane in each road segment in the next unit time based on the average vehicle speed of each lane in each road segment in the current unit time and a linear regression model. Then, based on preset lane selection conditions, the road segment length of each road segment, and the average vehicle speed of each lane in each road segment in the next unit time, the shortest time to reach and pass through the last road segment from the current position is calculated. Finally, the lane sequence corresponding to the shortest time is determined as the target lane sequence required for the vehicle to pass through the road segment to be traversed. Therefore, the embodiments of the present invention can not only plan the target lane sequence with the shortest travel time, thereby improving traffic efficiency, but also, when calculating the target lane sequence, can use a linear regression model trained based on historical average vehicle traffic speed to predict the average vehicle traffic speed of each lane in each road segment in the next unit of time. This makes the predicted average vehicle traffic speed of each lane in each road segment in the next unit of time closer to the actual average vehicle traffic speed. Consequently, the accuracy of selecting the target lane sequence with the shortest travel time based on the lane-level traffic conditions of each road segment in the next unit of time is higher.

[0073] Furthermore, the technical effects that can be achieved by the embodiments of the present invention include:

[0074] 1. By dividing roads based on historical average vehicle speeds, the frequency of lane changes can be kept moderate. This avoids the inefficiency of lane changes preventing the lane selection algorithm from fully improving speed, and also avoids the reduction in driving safety due to excessively high lane changes.

[0075] 2. When the target lane of the starting road segment and / or the last road segment of the target road is a fixed lane, in order to avoid lane planning failure, the estimated time of the other lanes of the starting road segment and the last road segment can be set to a preset estimated time that is greater than the estimated time of all lanes, so that the target lane sequence includes the fixed lane.

[0076] 3. To avoid sudden traffic congestion caused by abnormal traffic events, or significant changes in future traffic congestion even without abnormal traffic events, which would result in longer travel times when following the original target lane sequence, this embodiment of the invention can re-plan the lane-level traffic flow of the road segment to be traversed when an abnormal traffic event is detected or when the lane planning update time is reached, thereby improving traffic efficiency.

[0077] Of course, implementing any product or method of the present invention does not necessarily require achieving all of the advantages described above at the same time. Attached Figure Description

[0078] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the accompanying drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are merely some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without any creative effort.

[0079] Figure 1 A flowchart illustrating a lane selection method based on autonomous driving, provided as an embodiment of the present invention;

[0080] Figure 2 An example diagram illustrating a lane selection method based on autonomous driving provided in an embodiment of the present invention;

[0081] Figure 3 A flowchart illustrating another lane selection method based on autonomous driving provided in an embodiment of the present invention;

[0082] Figure 4 This is a block diagram of a lane selection device based on autonomous driving, provided as an embodiment of the present invention. Detailed Implementation

[0083] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0084] It should be noted that the terms "comprising" and "having," and any variations thereof, in the embodiments and drawings of this invention are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or device that includes a series of steps or units is not limited to the steps or units listed, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to these processes, methods, products, or devices.

[0085] This invention provides a lane selection method and apparatus based on autonomous driving, capable of pre-planning a target lane sequence with the shortest travel time, so as to perform autonomous driving based on the target lane sequence, thereby improving traffic efficiency. The method provided in this invention can be applied to any electronic device with computing power, such as an autonomous vehicle. In one implementation, the functional software implementing this method can exist as a standalone client software or as a plugin for existing related client software.

[0086] The embodiments of the present invention will be described in detail below.

[0087] Figure 1 This is a schematic flowchart of a lane selection method based on autonomous driving provided by an embodiment of the present invention. The method may include the following steps:

[0088] S100: Divide the target road into multiple road segments.

[0089] Specifically, the target road can be divided into multiple road segments of equal length, but this method is limited by the target road itself. This method is applicable to straight target roads of equal length (like highways) without intersections. However, when the target road includes intersections and / or has curves, this method may result in a suboptimal lane sequence, potentially even posing a risk of unsafe driving. Since the purpose of dividing the road into segments in this embodiment is to ensure that vehicles only consider lane changes when entering the next segment after completing one, the length of the road segment determines the lane-changing frequency. Dividing the target road into multiple road segments of equal length would result in road segments with different vehicle speeds having the same lane-changing frequency, thus failing to meet the different lane-changing needs of road segments with varying vehicle speeds.

[0090] To avoid the above-mentioned problems, this invention provides a target road division method applicable to various types of roads. The method includes: dividing the target road into multiple road segments based on the historical average vehicle speed on the target road, such that the historical average vehicle speed is positively correlated with the road segment length within a preset road segment length range.

[0091] Specifically, when the historical average vehicle speed is less than or equal to a first average vehicle speed threshold, the corresponding road segment length is a first length; when the historical average vehicle speed is greater than the first average vehicle speed threshold and less than a second average vehicle speed threshold, the corresponding road segment length is proportional to the historical average vehicle speed; when the historical average vehicle speed is greater than the second average vehicle speed threshold, the corresponding road segment length is a second length; wherein, the first average vehicle speed threshold is less than the second average vehicle speed threshold, and the first length is less than the second length.

[0092] In practical implementation, the division can be based on the following formula:

[0093]

[0094] in, The historical average vehicle speed is represented by h (in hours) and l (in seconds). seg The length of the road segment is indicated by 40 km / h as the first average vehicle speed threshold, 120 km / h as the second average vehicle speed threshold, 0.4 km as the first length, and 1.2 km as the second length.

[0095] Therefore, it can be seen that the method of dividing roads by historical average vehicle speed can ensure that the frequency of lane changing is moderate. This avoids the inability of lane selection algorithms to fully improve traffic speed due to excessively low lane changing frequency, and also avoids the reduction of vehicle driving safety due to excessively high lane changing frequency.

[0096] It should be noted that the target road is the road between the starting point and the destination set by the driver when the vehicle's autonomous driving function is activated. The target road includes one or more roads, and the specific presentation form of the target road is not limited in this embodiment of the invention. The historical average vehicle speed is calculated based on the historical instantaneous vehicle speed and the number of vehicles. The historical instantaneous vehicle speed is calculated based on the historical instantaneous vehicle speed collected by the roadside sensing device. For details, please refer to the explanation of instantaneous vehicle speed in step S110.

[0097] S110: Get the average vehicle speed of each lane of each road segment to be traversed in the current unit of time.

[0098] Specifically, the instantaneous vehicle speeds of all vehicles in each lane of each road segment to be traversed can be obtained from the roadside sensing device within the current unit of time. For each lane of each road segment to be traversed, the average vehicle speed of each lane within the current unit of time is calculated based on the number of vehicles at each sampling moment, the number of samplings within the current unit of time, and the instantaneous vehicle speed.

[0099] The unit time can be determined according to the actual situation; for example, the unit time can be 1 minute. For each lane of each road segment to be traversed, the average vehicle speed of each lane in the current unit time is calculated based on the number of vehicles at each sampling moment in the current unit time, the number of samplings in the current unit time, and the instantaneous vehicle speed. Specifically, this can be achieved by: for each lane of each road segment to be traversed, calculating the instantaneous vehicle speed at each sampling moment in the current unit time based on the number of vehicles at each sampling moment in the current unit time and the instantaneous vehicle speed; and for each lane of each road segment to be traversed, calculating the average vehicle speed of each lane in the current unit time based on the instantaneous vehicle speed at each sampling moment in the current unit time and the number of samplings in the current unit time.

[0100] The instantaneous vehicle speed is the average instantaneous vehicle speed in one lane of a road segment at a given sampling time. The formula for calculating the instantaneous vehicle speed is as follows:

[0101]

[0102] Among them, v i v represents the instantaneous vehicle speed at the i-th sampling time. j Let n represent the instantaneous vehicle speed of the j-th vehicle collected at the i-th sampling time. veh This represents the number of vehicles collected at the i-th sampling time.

[0103] The formula for calculating average vehicle speed is:

[0104]

[0105] Among them, V i，k (n) represents the average vehicle speed in the i-th lane of the k-th road segment during the n-th unit of time, v i This represents the instantaneous vehicle speed at the i-th sampling time within the n-th unit of time, where n is the instantaneous vehicle speed. sampled This represents the number of samples taken in the nth unit of time.

[0106] S120: Based on the average vehicle speed of each lane of each road segment in the current unit of time and the linear regression model, determine the average vehicle speed of each lane of each road segment in the next unit of time.

[0107] The linear regression model is trained based on historical average vehicle speeds and is used to determine the average vehicle speed of each lane in each road segment within the next unit of time. The historical average vehicle speed is the average vehicle speed over historical unit of time. The average vehicle speed is the average of the instantaneous vehicle speeds in one lane of a road segment at multiple sampling times within a unit of time.

[0108] In practical applications, since the traffic conditions at a specific location are influenced by the traffic conditions upstream and downstream, the current traffic conditions of the (k-1)th and (k+1)th road segments will affect the future traffic conditions of the kth road segment. Furthermore, the traffic conditions of the ith lane are affected by the traffic conditions of the (i-1)th and (i+1)th lanes. Therefore, the average vehicle speed V of the ith lane in the kth road segment during the (n+1)th unit of time can be obtained by training on historical average vehicle speeds. i，k (n+1), and the average vehicle speed V on the i-th lane of the k-th road segment within the n-th unit of time. i，k (n) and the average vehicle speed V at adjacent locations i-1，k-1 (n) to V i+1，k+1 The relationship between (n).

[0109] Specifically, the average vehicle speed V of the i-th lane in the k-th road segment during the (n+1)-th unit time can be calculated based on the linear regression model. i，k (n+1);

[0110] The linear regression model is as follows:

[0111] V i，k (n+1)=u+[a1 a2 a3 a4 a5 a6 a7 a8 a9]×[V i-1，k-1 (n) V i-1，k (n)V i-1，k+1 (n) V i，k-1 (n) V i，k (n) V i，k+1 (n) V i+1，k-1 (n) V i+1，k (n) V i+1，k+1 (n)] T

[0112] Where u, a1, a2, a3, a4, a5, a6, a7, a8, and a9 are the parameters of the linear regression model, and V i-1，k-1 (n) represents the average vehicle speed of the (i-1)th lane of the (k-1)th road segment in the nth unit of time, V i-1，k (n) represents the average vehicle speed of the (i-1)th lane of the k-th road segment in the nth unit of time. i-1，k+1 (n) represents the average vehicle speed of the (i-1)th lane in the (k+1)th road segment during the nth unit of time, V i，k-1 (n) represents the average vehicle speed of the i-th lane in the (k-1)-th road segment during the n-th unit of time, V i，k (n) represents the average vehicle speed of the i-th lane in the k-th road segment during the n-th unit of time, V i，k+1 (n) represents the average vehicle speed of the i-th lane in the (k+1)-th road segment during the n-th unit of time, V i+1，k-1 (n) represents the average vehicle speed of the (i+1)th lane of the (k-1)th road segment in the nth unit of time, V i+1，k (n) represents the average vehicle speed of the (i+1)th lane of the k-th road segment in the n-th unit of time, V i+1，k+1 (n) represents the average vehicle speed of the (i+1)th lane of the (k+1)th road segment in the nth unit of time.

[0113] S130: Based on preset lane selection conditions, the length of each road segment, and the average vehicle speed of each lane in each road segment in the next unit time, calculate the shortest time to reach and pass through the last road segment from the current position, and determine the lane sequence corresponding to the shortest time as the target lane sequence required for the vehicle to pass through the road segment to be passed.

[0114] Since frequent or continuous lane changes can easily affect driving safety, this embodiment of the invention allows lane changes between adjacent lanes. That is, the preset lane selection condition is that when moving from the current lane of the current road segment to the next road segment, the selectable lanes include the current lane and the lane adjacent to the current lane.

[0115] To reduce vehicle travel time, the shortest time to reach and pass through the last road segment from the current location can be calculated, provided the preset lane selection conditions are met. The lane sequence corresponding to the shortest time is then determined as the target lane sequence. The method for calculating the shortest time can be as follows: based on the road segment length and the average vehicle speed of each lane in each road segment within the next unit of time, the estimated time for a vehicle to pass through each lane in each road segment is calculated; based on the preset lane selection conditions and the estimated time for passing through each lane in each road segment, the shortest time to reach and pass through the last road segment from the current location is calculated.

[0116] Wherein, the estimated time T for passing through the i-th lane of the k-th road segment. i，k =l seg / V i，k , where l seg V is the length of the k-th road segment. i，k Let be the average vehicle speed of the i-th lane in the k-th road segment during the next unit of time.

[0117] There are three ways to reach and pass through the i-th lane of the k-th road segment: from the (i-1)-th lane of the (k-1)-th road segment to the i-th lane of the k-th road segment, from the i-th lane of the (k-1)-th road segment to the i-th lane of the k-th road segment, or from the (i+1)-th lane of the (k-1)-th road segment to the i-th lane of the k-th road segment. Therefore, the shortest time to reach and pass through the i-th lane of the k-th road segment is the minimum of the shortest times to reach and pass through the (i-1)-th lane of the (k-1)-th road segment, the shortest times to reach and pass through the i-th lane of the (k-1)-th road segment, and the shortest times to reach and pass through the (i+1)-th lane of the (k-1)-th road segment, plus the estimated time to reach and pass through the i-th lane of the k-th road segment.

[0118] Therefore, the method for calculating the shortest time from the current position to the last road segment can be as follows: iterative calculation is performed based on a preset formula and the estimated time for each lane of each road segment to obtain the shortest time from the current position to the last road segment for each lane, and the minimum value among the shortest times for each lane is determined as the shortest time to the last road segment.

[0119] Wherein, the preset formula is

[0120] S i，k =min{S i-1，k-1 S i，k-1 S i+1，k-1}+T i，k ,

[0121] Among them, S i，k S is the shortest time to reach and pass through the i-th lane of the k-th road segment from the current position. i-1，k-1 S is the shortest time to reach and pass through the (i-1)th lane of the (k-1)th road segment from the current position. i，k-1 S is the shortest time to reach and pass through the i-th lane of the (k-1)-th road segment from the current position. i+1，k-1 T is the shortest time to reach and pass through the (i+1)th lane of the (k-1)th road segment from the current position. i，k The estimated time to pass through the i-th lane of the k-th road segment.

[0122] For example, when the target lane is divided into 6 road segments, each road segment includes 4 lanes, after calculating the estimated time to pass through each lane of each road segment (see Table 1), the shortest time to reach and pass through each lane of each road segment can be determined under the condition of satisfying preset lane selection conditions (see Table 2). The shortest time to reach and pass through the 6th road segment from the 1st road segment is 11 minutes, and the corresponding target lane sequence (see the cells with gray backgrounds in Table 2) is: Lane 1 of the 1st road segment, Lane 2 of the 2nd road segment, Lane 3 of the 3rd road segment, Lane 4 of the 4th road segment, Lane 4 of the 5th road segment, and Lane 3 of the 6th road segment. Furthermore, Figure 2 The selected target lane sequence is visually displayed. The figure shows the average vehicle speed of each lane in each road segment within the next unit of time. The higher the speed, the lighter the color of the road segment. Under the premise of meeting the preset lane selection conditions, the target lane sequence selected in this embodiment of the invention plans a lane sequence with a relatively high average vehicle speed. In the figure, road segments 1-6 refer to the first to sixth road segments mentioned in this embodiment of the invention, and lanes 1-4 refer to the first to fourth lanes mentioned in this embodiment of the invention.

[0123] Table 1

[0124]

[0125] Table 2

[0126]

[0127] The lane selection method based on autonomous driving provided in this invention first divides the target road into multiple road segments, then obtains the average vehicle speed of each lane in each road segment to be traversed in the current unit time, and predicts the average vehicle speed of each lane in each road segment in the next unit time based on the average vehicle speed of each lane in each road segment in the current unit time and a linear regression model. Then, based on preset lane selection conditions, the road segment length of each road segment, and the average vehicle speed of each lane in each road segment in the next unit time, the shortest time to reach and pass through the last road segment from the current position is calculated. Finally, the lane sequence corresponding to the shortest time is determined as the target lane sequence required for the vehicle to pass through the road segment to be traversed. Therefore, the embodiments of the present invention can not only plan the target lane sequence with the shortest travel time, thereby improving traffic efficiency, but also, when calculating the target lane sequence, can use a linear regression model trained based on historical average vehicle traffic speed to predict the average vehicle traffic speed of each lane in each road segment in the next unit of time. This makes the predicted average vehicle traffic speed of each lane in each road segment in the next unit of time closer to the actual average vehicle traffic speed. Consequently, the accuracy of selecting the target lane sequence with the shortest travel time based on the lane-level traffic conditions of each road segment in the next unit of time is higher.

[0128] Optionally, in practical applications, the target lanes of the starting and / or last road segments of the target road that the driver wants to travel on are often fixed lanes. For example, the destination the driver wants to go to is often located on the edge of the road, meaning the lane in the last road segment is often an edge lane. In this case, if the target lane sequence planned by the autonomous vehicle does not include fixed lanes, it will cause the driver to be unable to reach the destination. To avoid this technical problem, when the target lane of the starting road segment is a fixed lane, the estimated time for the other lanes of the starting road segment can be set to a preset estimated time, which is greater than the estimated time of all lanes; when the target lane of the last road segment is a fixed lane, the estimated time for the other lanes of the last road segment can be set to the preset estimated time. The preset estimated time can be set to a value much larger than the historical estimated time, for example, 10. 6 Minutes. The target lane is the lane that needs to be actually driven in, that is, the lane that needs to be planned into the target lane sequence.

[0129] For example, assuming the target lane for the first road segment in Table 1 is a fixed lane, such as lane 2, and the target lane for the sixth road segment is a fixed lane, such as lane 4, then lanes 1, 3, and 4 of the first road segment and lanes 1, 2, and 3 of the sixth road segment in Table 1 need to be set to a preset estimated time, for example, 106 minutes. The estimated time for each lane in each road segment is detailed in Table 3, and the shortest time to reach and pass through each lane in each road segment is detailed in Table 4. Specifically, the shortest time to reach and pass through the sixth road segment from the first road segment is 14 minutes, and the corresponding target lane sequence (see the cells with gray backgrounds in Table 2) is: lane 2 of the first road segment, lane 2 of the second road segment, lane 3 of the third road segment, lane 4 of the fourth road segment, lane 4 of the fifth road segment, and lane 4 of the sixth road segment.

[0130] Table 3

[0131]

[0132] Table 4

[0133]

[0134] Optionally, since traffic congestion may occur during driving due to traffic anomalies (such as traffic accidents, sudden road closures due to weather), and traffic congestion may change significantly even without traffic anomalies, it may take a considerable amount of time if the entire journey is driven according to the initially planned target lane sequence. To further improve traffic efficiency, during driving based on the target lane sequence, the autonomous vehicle can detect in real time or periodically whether a traffic anomaly has occurred or whether the lane planning update time has been reached. When a traffic anomaly is detected or the lane planning update time has been reached, the vehicle returns to step S110, which involves obtaining the average vehicle speed of each lane in each road segment to be traversed within the current unit of time, in order to obtain the updated target lane sequence, and continues driving based on the updated target lane sequence. The lane planning update period corresponding to the lane planning update time can be one or more unit times mentioned in the embodiments of this invention.

[0135] The complete planning process for implementing this invention is as follows: Figure 3 As shown:

[0136] S200 divides the target road into multiple road segments.

[0137] S210. Obtain the average vehicle speed of each lane in each road segment to be traversed within the current unit of time.

[0138] S220. Based on the average vehicle speed of each lane in each road segment in the current unit of time and the linear regression model, determine the average vehicle speed of each lane in each road segment in the next unit of time.

[0139] The linear regression model is a model trained based on historical average vehicle speeds, used to determine the average vehicle speed of each lane in each road segment in the next unit of time. The historical average vehicle speed is the average vehicle speed in historical units of time.

[0140] S230. Based on preset lane selection conditions, the length of each road segment, and the average vehicle speed of each lane in each road segment in the next unit time, calculate the shortest time to reach and pass through the last road segment from the current position, and determine the lane sequence corresponding to the shortest time as the target lane sequence required for the vehicle to pass through the road segment to be passed.

[0141] The preset lane selection condition is that when moving from the current lane of the current road segment to the next road segment, the selectable lanes include the current lane and the lane adjacent to the current lane.

[0142] S240, Drive according to the target lane sequence.

[0143] S250: Detect whether a traffic anomaly has occurred or the lane planning update time has arrived; if a traffic anomaly is detected or the lane planning update time has arrived, return to execute S210; otherwise, continue to execute S240.

[0144] Based on the above method embodiments, this invention provides a lane selection device based on autonomous driving, such as... Figure 4 As shown, the device includes:

[0145] Division unit 30 is used to divide the target road into multiple road segments;

[0146] The acquisition unit 32 is used to acquire the average vehicle speed of each lane of each road segment to be traversed in the current unit of time.

[0147] The first determining unit 34 is used to determine the average vehicle speed of each lane in each road segment in the next unit time based on the average vehicle speed of each lane in each road segment in the current unit time and the linear regression model. The linear regression model is a model trained based on the historical average vehicle speed and used to determine the average vehicle speed of each lane in each road segment in the next unit time. The historical average vehicle speed is the average vehicle speed in a historical unit time.

[0148] The calculation unit 36 ​​is used to calculate the shortest time to reach and pass through the last road segment from the current position based on preset lane selection conditions, the road segment length of each road segment, and the average vehicle speed of each lane of each road segment in the next unit of time.

[0149] The second determining unit 38 is used to determine the lane sequence corresponding to the shortest time as the target lane sequence required for the vehicle to pass through the road segment to be passed. The preset lane selection condition is that when moving from the current lane of the current road segment to the next road segment, the selectable lanes include the current lane and the lane adjacent to the current lane.

[0150] Optionally, the division unit 30 is used to divide the target road into multiple road segments based on the historical average vehicle speed on the target road, such that the historical average vehicle speed is positively correlated with the road segment length within a preset road segment length range.

[0151] Optionally, the division unit 30 is used to determine the first length of the corresponding road segment when the historical average vehicle speed is less than or equal to the first average vehicle speed threshold.

[0152] When the historical average vehicle speed is greater than the first average vehicle speed threshold and less than the second average vehicle speed threshold, the corresponding road segment length is proportional to the historical average vehicle speed.

[0153] When the historical average vehicle speed is greater than the second average vehicle speed threshold, the corresponding road segment length is the second length.

[0154] Wherein, the first average vehicle speed threshold is less than the second average vehicle speed threshold, and the first length is less than the second length.

[0155] Optionally, the acquisition unit 32 includes:

[0156] The acquisition module is used to acquire the instantaneous vehicle speeds of all vehicles in each lane of each road segment to be traversed, collected by the roadside sensing device within the current unit of time.

[0157] The first calculation module is used to calculate the average vehicle speed of each lane in the current unit time for each lane of each road segment to be traversed, based on the number of vehicles at each sampling moment in the current unit time, the number of samplings in the current unit time, and the instantaneous vehicle speed.

[0158] Optionally, the first determining unit 34 is used to calculate the average vehicle speed V of the i-th lane of the k-th road segment in the (n+1)-th unit time according to the first formula. i，k (n+1);

[0159] The first formula is

[0160] V i，k (n+1)=u+[a1 a2 a3 a4 a5 a6 a7 a8 a9]×[V i-1，k-1 (n) V i-1，k (n)V i-1，k+1 (n) V i，k-1 (n) V i，k (n) V i，k+1 (n) V i+1，k-1 (n) V i+1，k (n) V i+1，k+1 (n)] T

[0161] Where u, a1, a2, a3, a4, a5, a6, a7, a8, and a9 are the parameters of the linear regression model, and V i-1，k-1 (n) represents the average vehicle speed of the (i-1)th lane of the (k-1)th road segment in the nth unit of time, V i-1，k (n) represents the average vehicle speed of the (i-1)th lane of the k-th road segment in the nth unit of time. i-1，k+1 (n) represents the average vehicle speed of the (i-1)th lane in the (k+1)th road segment during the nth unit of time, V i，k-1 (n) represents the average vehicle speed of the i-th lane in the (k-1)-th road segment during the n-th unit of time, V i，k (n) represents the average vehicle speed of the i-th lane in the k-th road segment during the n-th unit of time, V i，k+1 (n) represents the average vehicle speed of the i-th lane in the (k+1)-th road segment during the n-th unit of time, V i+1，k-1 (n) represents the average vehicle speed V of the (i+1)th lane of the (k-1)th road segment in the nth unit of time. i+1，k (n) represents the average vehicle speed of the (i+1)th lane of the k-th road segment in the n-th unit of time, V i+1，k+1 (n) represents the average vehicle speed of the (i+1)th lane of the (k+1)th road segment in the nth unit of time.

[0162] Optionally, the computing unit 36 ​​includes:

[0163] The second calculation module is used to calculate the estimated time for a vehicle to pass through each lane of each road segment based on the road segment length of each road segment and the average vehicle speed of each lane of each road segment in the next unit of time.

[0164] The third calculation module is used to calculate the shortest time from the current position to the last road segment based on the preset lane selection conditions and the estimated time to pass through each lane of each road segment.

[0165] Optionally, a third calculation module is used to iteratively calculate based on a preset formula and the estimated time for each lane of each road segment to obtain the shortest time from the current position to reach and pass through each lane of the last road segment, and to determine the minimum value among the shortest times of each lane as the shortest time to reach and pass through the last road segment.

[0166] Wherein, the preset formula is

[0167] S i，k =min{S i-1，k-1 S i，k-1 S i+1，k-1}+T i，k ,

[0168] Among them, S i，k S is the shortest time to reach and pass through the i-th lane of the k-th road segment from the current position. i-1，k-1 S is the shortest time to reach and pass through the (i-1)th lane of the (k-1)th road segment from the current position. i，k-1 S is the shortest time to reach and pass through the i-th lane of the (k-1)-th road segment from the current position. i+1，k-1 T is the shortest time to reach and pass through the (i+1)th lane of the (k-1)th road segment from the current position. i，k The estimated time to pass through the i-th lane of the k-th road segment.

[0169] Optionally, the device further includes:

[0170] The setting unit is configured to, when the target lane of the starting road segment is a fixed lane, set the estimated time of the other lanes of the starting road segment to a preset estimated time, wherein the preset estimated time is greater than the estimated time of all lanes; and when the target lane of the last road segment is a fixed lane, set the estimated time of the other lanes of the last road segment to the preset estimated time.

[0171] Optionally, the device further includes:

[0172] The return execution unit is used to, during the driving process based on the target lane sequence, when a traffic anomaly event is detected or the lane planning update time is reached, return to execute the step of obtaining the average vehicle speed of each lane of each road segment to be passed in the current unit time, so as to obtain the updated target lane sequence, and continue driving based on the updated target lane sequence.

[0173] Based on the above method embodiments, another embodiment of the present invention provides a storage medium storing executable instructions that, when executed by a processor, cause the processor to implement the method described above.

[0174] Based on the above method embodiments, another embodiment of the present invention provides an autonomous driving vehicle, the autonomous driving vehicle comprising:

[0175] One or more processors;

[0176] Storage device for storing one or more programs.

[0177] When the one or more programs are executed by the one or more processors, the one or more processors implement the method described above.

[0178] The above system and device embodiments correspond to the method embodiments and have the same technical effects. For detailed descriptions, please refer to the method embodiments. The device embodiments are derived based on the method embodiments; detailed descriptions can be found in the method embodiments section, and will not be repeated here. Those skilled in the art will understand that the accompanying drawings are merely schematic diagrams of one embodiment, and the modules or processes shown in the drawings are not necessarily essential for implementing the present invention.

[0179] Those skilled in the art will understand that the modules in the apparatus of the embodiments can be distributed in the apparatus of the embodiments as described in the embodiments, or they can be located in one or more devices different from this embodiment with corresponding changes. The modules of the above embodiments can be combined into one module, or they can be further divided into multiple sub-modules.

[0180] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A lane selection method based on autonomous driving, characterized in that, The method includes: Divide the target road into multiple road segments; Get the average vehicle speed of each lane in each road segment to be traversed in the current unit of time; Based on the average vehicle speed of each lane in each road segment in the current unit time and the linear regression model, the average vehicle speed of each lane in each road segment in the next unit time is determined. The linear regression model is a model trained based on the historical average vehicle speed and used to determine the average vehicle speed of each lane in each road segment in the next unit time. The historical average vehicle speed is the average vehicle speed in a historical unit time. Based on preset lane selection conditions, the length of each road segment, and the average vehicle speed of each lane in each road segment in the next unit time, the shortest time to reach and pass through the last road segment from the current position is calculated, and the lane sequence corresponding to the shortest time is determined as the target lane sequence required for the vehicle to pass through the road segment to be passed. The preset lane selection conditions are that when moving from the current lane of the current road segment to the next road segment, the selectable lanes include the current lane and the lane adjacent to the current lane. The target road is divided into multiple road segments, including: When the historical average vehicle speed on the target road is less than or equal to the first average vehicle speed threshold, the corresponding road segment length is the first length. When the historical average vehicle speed is greater than the first average vehicle speed threshold and less than the second average vehicle speed threshold, the corresponding road segment length is proportional to the historical average vehicle speed. When the historical average vehicle speed is greater than the second average vehicle speed threshold, the corresponding road segment length is the second length. Wherein, the first average vehicle speed threshold is less than the second average vehicle speed threshold, and the first length is less than the second length.

2. The method as described in claim 1, characterized in that, Obtain the average vehicle speed for each lane of each road segment to be traversed within the current unit of time, including: The instantaneous vehicle speeds of all vehicles in each lane of each road segment to be traversed, collected by roadside sensing devices, are obtained within the current unit of time. For each lane of each road segment to be traversed, the average vehicle speed of each lane in the current unit of time is calculated based on the number of vehicles at each sampling moment in the current unit of time, the number of samplings in the current unit of time, and the instantaneous vehicle speed.

3. The method as described in claim 1, characterized in that, Based on the average vehicle speed of each lane in each road segment in the current unit of time and a linear regression model, determine the average vehicle speed of each lane in each road segment in the next unit of time, including: The average vehicle speed of the i-th lane on the k-th road segment in the (n+1)-th unit time is calculated based on the linear regression model. ; The linear regression model is as follows: ； in, , , , , , , , and These are the parameters of the linear regression model. Let be the average vehicle speed of the (i-1)th lane in the (k-1)th road segment during the nth unit of time. Let be the average vehicle speed of the (i-1)th lane of the k-th road segment in the nth unit of time. Let be the average vehicle speed of the (i-1)th lane in the (k+1)th road segment during the nth unit of time. Let be the average vehicle speed of the i-th lane in the (k-1)-th road segment during the n-th unit of time. Let be the average vehicle speed of the i-th lane in the k-th road segment during the n-th unit of time. Let be the average vehicle speed of the i-th lane in the (k+1)-th road segment during the n-th unit of time. Let be the average vehicle speed of the (i+1)th lane in the (k-1)th road segment during the nth unit of time. Let be the average vehicle speed of the (i+1)th lane of the k-th road segment in the n-th unit of time. Let be the average vehicle speed of the (i+1)th lane of the (k+1)th road segment in the nth unit of time.

4. The method as described in claim 1, characterized in that, Based on preset lane selection conditions, the length of each road segment, and the average vehicle speed of each lane in each road segment within the next unit of time, calculate the shortest time to reach and pass through the last road segment from the current location, including: Based on the road segment length of each road segment and the average vehicle speed of each lane in each road segment in the next unit of time, calculate the estimated time for a vehicle to pass through each lane of each road segment. Based on the preset lane selection conditions and the estimated time to pass through each lane of each road segment, calculate the shortest time to reach and pass through the last road segment from the current location.

5. The method as described in claim 4, characterized in that, Based on the preset lane selection conditions and the estimated time for traversing each lane of each road segment, calculate the shortest time from the current location to and through the last road segment, including: Based on the preset formula and the estimated time for each lane of each road segment, the shortest time to reach and pass through the last road segment from the current position is obtained for each lane. The minimum value among the shortest times for each lane is determined as the shortest time to reach and pass through the last road segment. Wherein, the preset formula is ， in, The shortest time to reach and pass through the i-th lane of the k-th road segment from the current position. The shortest time to reach and pass through the (i-1)th lane of the (k-1)th road segment from the current position. The shortest time to reach and pass through the i-th lane of the (k-1)-th road segment from the current position. The shortest time to reach and pass through the (i+1)th lane of the (k-1)th road segment from the current position. The estimated time to pass through the i-th lane of the k-th road segment.

6. The method as described in claim 1, characterized in that, The method further includes: When the target lane of the starting road segment is a fixed lane, the estimated time of the other lanes of the starting road segment is set to a preset estimated time, which is greater than the estimated time of all lanes. When the target lane of the last road segment is a fixed lane, the estimated time of the other lanes of the last road segment is set to the preset estimated time.

7. The method according to any one of claims 1-6, characterized in that, The method further includes: During the driving process based on the target lane sequence, when a traffic anomaly is detected or the lane planning update time is reached, the process returns to obtain the average vehicle speed of each lane of each road segment to be traversed in the current unit of time, so as to obtain the updated target lane sequence, and continues driving based on the updated target lane sequence.

8. A lane selection device based on autonomous driving, characterized in that, The device includes: Dividing units are used to divide a target road into multiple road segments; The acquisition unit is used to acquire the average vehicle speed of each lane in each road segment to be traversed in the current unit of time. The first determining unit is used to determine the average vehicle speed of each lane of each road segment in the next unit time based on the average vehicle speed of each lane of each road segment in the current unit time and a linear regression model. The linear regression model is a model trained based on historical average vehicle speeds to determine the average vehicle speed of each lane of each road segment in the next unit time. The historical average vehicle speed is the average vehicle speed in historical unit time. The calculation unit is used to calculate the shortest time to reach and pass through the last road segment from the current location based on preset lane selection conditions, the road segment length of each road segment, and the average vehicle speed of each lane of each road segment in the next unit of time. The second determining unit is used to determine the lane sequence corresponding to the shortest time as the target lane sequence required for the vehicle to pass through the road segment to be passed. The preset lane selection condition is that when moving from the current lane of the current road segment to the next road segment, the selectable lanes include the current lane and the lane adjacent to the current lane. The division unit is used to determine the first length of the corresponding road segment when the historical average vehicle speed on the target road is less than or equal to the first average vehicle speed threshold. When the historical average vehicle speed is greater than the first average vehicle speed threshold and less than the second average vehicle speed threshold, the corresponding road segment length is proportional to the historical average vehicle speed. When the historical average vehicle speed is greater than the second average vehicle speed threshold, the corresponding road segment length is the second length. Wherein, the first average vehicle speed threshold is less than the second average vehicle speed threshold, and the first length is less than the second length.

Images