A left-turn path control method and device, vehicle and storage medium

Abstract

Embodiments of the present application relate to the technical field of automatic driving, and disclose a left-turn path control method and device, a vehicle and a storage medium. The method comprises the following steps: acquiring a target driving direction of a self-vehicle at a next intersection, and determining a target lane corresponding to a left turn; acquiring a lane length of the target lane, the lane length being a length of the target lane between a current intersection and the next intersection; and planning a path of the self-vehicle from a current lane at the current intersection to the target lane based on a size relationship between the lane length of the target lane and a passing threshold. By using the technical solution of the present application, the selection error of the target lane in the passable lane can be avoided, the risk of left-turn failure can be reduced, and the overall passing efficiency is improved.

Description

Technical Field

[0001] This invention relates to the field of autonomous driving technology, specifically to a left-turn path control method, device, vehicle, and storage medium. Background Technology

[0002] In related technologies, left-turn control in Advanced Driver Assistance Systems (ADAS) requires multiple lane changes to stay in the correct lane. Summary of the Invention

[0003] In view of the above problems, embodiments of the present invention provide a left-turn path control method, device, vehicle and storage medium to solve the problem in the prior art that a vehicle needs to change lanes multiple times to travel in the correct lane when turning left.

[0004] According to one aspect of the present invention, a left-turn path control method is provided, applied to a vehicle, the method comprising: obtaining the target driving direction of the vehicle at the next intersection, and determining the target lane corresponding to the left turn; obtaining the lane length of the target lane, the lane length being the length of the target lane between the current intersection and the next intersection; and planning the path for the vehicle to turn left from the current lane at the current intersection to the target lane based on the relationship between the lane length of the target lane and a traffic threshold.

[0005] According to another aspect of the present invention, a left-turn path control device is provided, comprising: a lane acquisition module, configured to acquire the target driving direction of a vehicle at the next intersection and determine the target lane corresponding to the left turn; a length acquisition module, configured to acquire the lane length of the target lane, wherein the lane length is the length of the target lane between the current intersection and the next intersection; and a path planning module, configured to plan the path for the vehicle to turn left from the current lane at the current intersection to the target lane based on the relationship between the lane length of the target lane and a passage threshold.

[0006] According to another aspect of the present invention, a vehicle is provided, comprising: a processor, a memory, a communication interface, and a communication bus, wherein the processor, the memory, and the communication interface communicate with each other via the communication bus; the memory is used to store at least one executable instruction, the executable instruction causing the processor to perform the operation of the aforementioned left-turn path control method.

[0007] According to another aspect of the present invention, a computer-readable storage medium is provided, the storage medium storing at least one executable instruction that causes a vehicle / device to perform the operation of the aforementioned left-turn path control method.

[0008] The technical solution provided by this invention obtains the target driving direction of the vehicle at the next intersection, determines the target lane corresponding to the left turn, and plans the path of the vehicle from the current lane at the current intersection to the target lane based on the relationship between the lane length of the target lane and the traffic threshold. This allows the vehicle to obtain the target driving direction of the next intersection in advance before turning left at the current intersection, thereby determining the target lane when the vehicle turns left at the current intersection and planning the path of the vehicle from the current lane at the current intersection to the target lane, thus improving the traffic efficiency of the vehicle at the next intersection after turning left.

[0009] The above description is merely an overview of the technical solutions of the embodiments of the present invention. In order to better understand the technical means of the embodiments of the present invention and to implement them in accordance with the contents of the specification, and to make the above and other objects, features and advantages of the embodiments of the present invention more apparent and understandable, specific embodiments of the present invention are described below. Attached Figure Description

[0010] The accompanying drawings are for illustrative purposes only and are not intended to limit the invention. Furthermore, the same reference numerals denote the same parts throughout the drawings. In the drawings:

[0011] Figure 1 A flowchart illustrating a left-turn path control method provided by the present invention is shown;

[0012] Figure 2 This invention provides another schematic flowchart of a left-turn path control method.

[0013] Figure 3 This invention provides another schematic flowchart of a left-turn path control method.

[0014] Figure 4 The diagram illustrates an application scenario of the left-turn path control method provided by this invention.

[0015] Figure 5 A schematic diagram of the structure of a left-turn path control device provided by the present invention is shown;

[0016] Figure 6 A schematic diagram of the structure of a vehicle provided by the present invention is shown;

[0017] Figure 7 Another structural schematic diagram of a vehicle provided by the present invention is shown;

[0018] Figure 8 Another structural schematic diagram of a vehicle provided by the present invention is shown. Detailed Implementation

[0019] Exemplary embodiments of the invention will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the invention are shown in the drawings, it should be understood that the invention can be implemented in various forms and should not be limited to the embodiments set forth herein.

[0020] Figure 1 A flowchart of a first embodiment of a left-turn path control method according to the present invention is shown, the method being executed by vehicle equipment. Figure 1 As shown, the method includes the following steps:

[0021] Step 110: Obtain the target driving direction of the vehicle at the next intersection and determine the target lane corresponding to the left turn.

[0022] In this embodiment of the invention, the target driving direction of the vehicle at the next intersection is determined by the target location to be reached and the locations along the route.

[0023] For example, when a vehicle determines the target lane for a left turn based on the target driving direction at the next intersection, if the target driving direction at the next intersection is a right turn, then the target lane corresponding to the vehicle turning left at the current intersection is a right turn lane; if the target driving direction at the next intersection is straight, then the target lane corresponding to the vehicle turning left at the current intersection is a straight lane.

[0024] Step 120: Obtain the lane length of the target lane.

[0025] The lane length is the length of the target lane between the current intersection and the next intersection.

[0026] In one alternative approach, when obtaining the lane length of the target lane, the vehicle can send a request to the big data traffic management platform to receive the lane length of the target lane from the big data traffic management platform, so as to plan the vehicle's left-turn path and improve the vehicle's traffic efficiency and reliability at the next intersection after the left turn.

[0027] In one alternative approach, when obtaining the lane length of the target lane, the vehicle can also accurately determine the lane length using its own forward-facing camera through image recognition, and plan its left-turn path based on the target lane length. Additionally, in some special scenarios, if the intersection sign itself has communication capabilities, it can establish a communication connection with the vehicle, allowing the vehicle to directly receive the target lane length sent by the intersection sign.

[0028] Step 130: Based on the relationship between the lane length of the target lane and the passage threshold, plan the path for the vehicle to turn left from the current lane at the current intersection to the target lane.

[0029] The passage threshold is determined based on congestion information between the current intersection and the next intersection.

[0030] In one alternative approach, when determining the passage threshold, the congestion information of the current road for the passable lanes can be obtained first. The passable lanes include the lanes between the current intersection and the next intersection. Then, the congestion level of the passable lanes is determined based on the congestion information. Finally, the passage threshold is determined based on the congestion level and the minimum lane change threshold, where the minimum lane change threshold is the minimum distance required for a vehicle to change lanes from the adjacent lane to the target lane after making a left turn.

[0031] Specifically, since the determination of the passage threshold Dmin depends on road-level congestion information, which can be obtained from the navigation system, after obtaining the congestion information of the passable lanes, the passage threshold Dmin can be interpolated based on the congestion information of the passable lanes. The minimum distance Dx required for a vehicle to change lanes from the adjacent lane to the target lane after turning left must also be considered, and the minimum distance Dx can be a fixed threshold value determined later. The relationship between congestion information and the congestion level of passable lanes is shown in Table 1.

[0032] Table 1. Relationship between congestion information and congestion level of passable lanes

[0033]

[0034] In the table, D1 > D2 > D3 > D4, meaning that the higher the congestion level, the larger the set passage threshold Dmin, and Dmin > Dx. In other embodiments, the congestion level classification can also be adjusted according to actual needs.

[0035] In one alternative approach, when planning the path for a vehicle to turn left from the current lane to the target lane at the current intersection, based on the relationship between the target lane length and the passage threshold, the lane length of the target lane and the passage threshold can be compared to obtain a threshold comparison result; then, based on the threshold comparison result, the path for the vehicle to change lanes from the current lane to the right turn lane can be planned.

[0036] Specifically, the length of the target lane can be compared with a passage threshold to determine if the target lane length is less than or equal to the passage threshold. If the target lane length is less than or equal to the passage threshold, it indicates that if the vehicle turns left at the current intersection into a lane that is not the target lane, there is insufficient space to change lanes to the target lane, thus preventing it from reaching the target direction at the next intersection. In this case, the path planning method is determined to be a direct left turn from the current lane to the target lane. The target lane can be unique (e.g., if the target direction is right, there is only one right-turn lane that the current lane can enter) or non-unique (e.g., if the target direction is right, there are multiple right-turn lanes that the current lane can enter). If the target lane is unique, the path for the vehicle to turn left from the current lane to the target lane is planned directly based on the target lane. If the target lane is not unique, the distance between each target lane and the target direction can be used to determine the path, i.e., the target lane closest to the target direction is selected first.

[0037] Secondly, the length of the target lane can be compared with the passage threshold to determine if the target lane length exceeds the threshold. If the target lane length exceeds the threshold, it indicates that if the vehicle turns left at the current intersection into a lane that is not the target lane, there is still enough space to change lanes to the target lane, thus changing lanes to the target lane corresponding to the target travel direction. At this point, the route planning methods include directly turning left to the target lane or turning left from the current lane to another lane and then changing lanes to the target lane. However, since there are many lane-changing methods, the optimal lane-changing method can be determined based on the scores of each method.

[0038] Compared to related technologies that require multiple lane changes to stay in the correct lane during left turn control, the left turn path control method of this invention obtains the target driving direction of the vehicle at the next intersection, determines the target lane for the left turn, and plans the path for the vehicle to turn left from the current lane to the target lane based on the relationship between the lane length of the target lane and the traffic threshold. This allows the vehicle to obtain the target driving direction of the next intersection in advance before turning left at the current intersection, thereby determining the target lane for the vehicle when turning left at the current intersection and planning the path for the vehicle to turn left from the current lane to the target lane, thus improving the traffic efficiency of the vehicle at the next intersection after turning left.

[0039] Figure 2 A flowchart of another embodiment of the left-turn path control method of the present invention is shown, which is executed by vehicle equipment. Figure 2 As shown, the method includes the following steps:

[0040] Step 210: Obtain the target driving direction of the vehicle at the next intersection and determine the target lane corresponding to the left turn.

[0041] Please see details Figure 1 Step 110 of the illustrated embodiment will not be described again here.

[0042] Step 220: Obtain the lane length of the target lane.

[0043] Please see details Figure 1 Step 120 of the illustrated embodiment will not be described again here.

[0044] Step 230: Based on the relationship between the lane length of the target lane and the passage threshold, plan the path for the vehicle to turn left from the current lane at the current intersection to the target lane.

[0045] Specifically, step 230 above includes:

[0046] Step 2301: If the length of the target lane is less than or equal to the passage threshold, the route planning method is determined to be a direct left turn from the current lane to the target lane, and it is determined whether the target lane is unique.

[0047] Specifically, if the length of the target lane is less than or equal to the traffic threshold, it means that if the vehicle turns left at the current intersection into a lane that is not the target lane, there is insufficient space to change lanes to the target lane, thus preventing it from reaching the target direction at the next intersection. Therefore, the path planning method is determined to be a direct left turn from the current lane to the target lane. Considering that the target lane may be unique or not, to improve the efficiency of the vehicle entering the target direction at the next intersection, it is necessary to determine the uniqueness of the target lane.

[0048] Step 2302: If the target lane is unique, then plan the path for the vehicle to turn left from the current lane to the target lane based on the path planning method.

[0049] Specifically, if there is only one target lane, the path for the vehicle to turn left from the current lane to the target lane is planned directly based on the target lane, thereby avoiding the vehicle driving in the left-turn lane. When the straight-ahead green light allows the vehicle to turn left into the waiting area, if the trajectory planning is based on the vehicle's current lane and the corresponding lane after turning left into the road, without selecting the target lane based on the target driving direction at the next intersection, and the left-turn path planning is based on the correspondence between the vehicle's current lane position and the target lane, then in some scenarios, the left turn may fail due to the large turning radius or insufficient passage space, or multiple lane changes may be required after turning left into the corresponding lane to drive in the target lane.

[0050] For example, if there is only one target lane, the vehicle will directly enter the lane corresponding to the target direction of travel. If the next intersection requires a right turn, the vehicle will enter the right turn lane of the passable lane (target road). If the next intersection requires a left turn, the vehicle will enter the left turn lane of the passable lane in advance. If the next intersection requires going straight, the target lane is the straight lane.

[0051] Step 2303: If the target lane is not unique, obtain the first path score of the vehicle turning left from the current lane to each target lane.

[0052] The first path score is determined based on each target lane and the target driving direction.

[0053] Specifically, if there is not a unique target lane, the system obtains the first path score for the vehicle to turn left from the current lane to each target lane, and then plans the vehicle's left-turn path based on the first path score of each target lane. When determining the first path score of each target lane, it can be based on the distance between each right-turn lane and the target direction of the next intersection, as well as the congestion information of each target lane; alternatively, it can be based solely on the distance between each right-turn lane and the target direction of the next intersection—that is, the closer the target lane is to the target direction of the next intersection, the higher its score—and there are no further restrictions on this.

[0054] For example, if the target driving direction is a right turn, and the right turn lanes that can be entered after turning left from the current lane are numbered 3, 2, and 1 from right to left, then the first path from the current lane to the target lane numbered 3 has the highest score, the first path from the current lane to the target lane numbered 2 has the second highest score, and the first path from the current lane to the target lane numbered 1 has the lowest score.

[0055] Step 2304: Based on the path planning method and the first path score, plan the path for the vehicle to turn left from the current lane to the target lane.

[0056] Specifically, by planning the path from the current lane to the target lane based on the path planning method and the first path score, the system can avoid situations where the vehicle is driving in the left-turn lane and can turn left into the waiting area when the straight-ahead green light is on. However, if the trajectory planning is based on the vehicle's current lane and the corresponding lane after turning left into the road, without selecting the target lane based on the target driving direction at the next intersection, and the left-turn path planning uses the correspondence between the vehicle's current lane position and the target lane for left-turn control, then in some scenarios, the left turn may fail due to the large turning radius or insufficient passage space, or multiple lane changes may be required after turning left into the corresponding lane to drive into the target lane.

[0057] In one alternative approach, if the path planning method involves turning left directly from the current lane to the target lane, then when planning the path for the vehicle to turn left from the current lane to the target lane, step 2301 specifically includes:

[0058] Step a1: Obtain the current position of the vehicle.

[0059] In this process, the current location of the vehicle is obtained to facilitate the planning of a left-turn route based on the vehicle's current location.

[0060] Step a2: Fit the current position of the vehicle to the center line of the target lane to obtain the fitting result.

[0061] The process involves fitting the vehicle's current position to the centerline of the target lane to obtain a fitting result, which provides the necessary conditions for the vehicle's path planning when turning left from the current lane to the target lane.

[0062] Step a3: Based on the fitting results, plan the path for the vehicle to turn left from the current lane to the target lane.

[0063] Specifically, by planning the path for the vehicle to turn left from the current lane to the target lane based on the fitting results, the target trajectory can be formed by fitting the vehicle's current position point and the center line of the target lane, rather than by fitting the current driving lane and the target lane. This means that the vehicle can drive along the lane lines instead of simply planning the trajectory according to the lane lines, thus avoiding situations where the turning radius is too large.

[0064] In this embodiment of the application, if the length of the target lane is less than or equal to the passage threshold, the path planning method is determined to be a direct left turn from the current lane to the target lane, and it is determined whether the target lane is unique. If the target lane is unique, the path of the vehicle turning left from the current lane to the target lane is planned based on the path planning method. If the target lane is not unique, the first path score of the vehicle turning left from the current lane to each target lane is obtained, thereby improving the passage efficiency of the vehicle at the next intersection after turning left.

[0065] Figure 3 A flowchart of another embodiment of the left-turn path control method of the present invention is shown, which is executed by vehicle equipment. Figure 3 As shown, the method includes the following steps:

[0066] Step 310: Obtain the target driving direction of the vehicle at the next intersection and determine the target lane corresponding to the left turn.

[0067] Please see details Figure 1 Step 110 of the illustrated embodiment will not be described again here.

[0068] Step 320: Obtain the lane length of the target lane.

[0069] Please see details Figure 1 Step 120 of the illustrated embodiment will not be described again here.

[0070] Step 330: Based on the relationship between the lane length of the target lane and the passage threshold, plan the path for the vehicle to turn left from the current lane at the current intersection to the target lane.

[0071] Specifically, step 330 above includes:

[0072] Step 3301: If the length of the target lane is greater than the passage threshold, the route planning method is determined to include turning left directly to the target lane and turning left from the current lane to another lane and then changing lanes to the target lane.

[0073] Specifically, if the length of the target lane is greater than the passage threshold, the route planning method is determined to include turning left directly to the target lane and turning left from the current lane to another lane and then changing lanes to the target lane. This means that if the vehicle turns left at the current intersection and the left-turn lane is not the target lane, there is still enough space to change lanes to the target lane, thus changing lanes to the target lane corresponding to the target driving direction. In other words, the target lane is not unique, meaning that the vehicle can enter any lane and then change lanes to the target lane. Therefore, when planning the left-turn route, the left-turn route can be planned as turning left directly to the target lane and turning left from the current lane to another lane and then changing lanes to the target lane.

[0074] Step 3302: Obtain the second path score for entering the lane after turning left for each path planning method.

[0075] Specifically, when planning the path for a vehicle to turn left from the current lane to the target lane at the current intersection, a second path score can be obtained, which can be either a direct left turn to the target lane or a left turn from the current lane to another lane and then a lane change to the target lane. The path for the vehicle to turn left from the current lane to the target lane can be planned based on the second path score.

[0076] Step 3303: Based on the path planning method and the second path score, plan the path for the vehicle to turn left from the current lane at the current intersection to the target lane.

[0077] Specifically, by planning the path for a vehicle to turn left from its current lane to the target lane at the current intersection based on path planning methods and second path scoring, it can avoid situations where the vehicle is driving in the left-turn lane and can turn left into the waiting area when the straight-ahead green light is on. However, if the trajectory planning is based on the vehicle's current lane and the corresponding lane after turning left into the road, without selecting the target lane based on the target driving direction at the next intersection, and if the left-turn path planning uses the correspondence between the vehicle's current lane position and the target lane for left-turn control, then in some scenarios, the left turn may fail due to an excessively large turning radius or insufficient passage space, or multiple lane changes may be required after turning left into the corresponding lane to reach the target lane.

[0078] In an alternative approach, step 3301 specifically includes:

[0079] Step b1: Obtain the path parameters and corresponding parameter weights of the vehicle changing lanes from the current intersection to each lane.

[0080] The path parameters include detour parameters for the vehicle turning left at the current intersection, parameters for failed turns, left turn passage time, and direct access parameters to the target lane.

[0081] Specifically, the detour parameters for a vehicle turning left at the current intersection are determined by the obstacles ahead on any path planned by the vehicle using path planning. This results in an evaluation result C1 value (the parameter value corresponding to the detour parameters), with higher values ​​for routes that do not require detours. In other words, the detour parameters are determined based on the number of times the vehicle needs to detour before turning left to a passable lane (fewer detours result in a higher C1 value), the obstacle detour difficulty coefficient (a smoother trajectory and a higher probability of successful detour result in a higher C1 value), and the type of obstacle to detour (stationary obstacles > moving small vehicles > moving large vehicles > two-wheeled vehicles > pedestrians).

[0082] The turning failure parameter is determined by detecting the turning radius before turning left into a passable lane, setting a minimum threshold R1 and a maximum threshold R2, and then writing these values ​​through subsequent real-vehicle calibration. The setting requires the radius to be within the range (R1, R2). A larger turning radius requires more turning space and increases the probability of failure; conversely, a smaller turning radius exceeds the vehicle's control limits and also increases the probability of failure. By calculating the turning radii of multiple trajectories, an evaluation result value C2 (the parameter value corresponding to the turning failure parameter) is obtained. When the turning radius is within the range (R1, R2), C2 takes a first fixed value (to be calibrated); when the turning radius is greater than R2, C2 takes a value of 0; and when the turning radius is less than R1, C2 takes a second fixed value.

[0083] The left-turn passage time is calculated by detecting information such as the path length and vehicle speed before turning left to the passable lane. By comparing the passage time, the evaluation result C3 value (the parameter value corresponding to the left-turn passage time) is obtained. The smaller the time, the higher the C3 value.

[0084] The direct access parameter for the target lane is determined by the distance between the vehicle after turning left from the current lane to the target lane and the target direction of travel at the next intersection. The closer the vehicle is to the target direction of travel at the next intersection, the better the target lane, meaning a higher C4 value (the parameter value corresponding to the direct access parameter). For example, if the target direction of travel is a right turn, and the right-turn lanes that can be entered after turning left from the current lane are numbered 3, 2, and 1 from right to left, then the first path from the current lane to target lane number 3 has the highest score, followed by the first path to target lane number 2, and the first path to target lane number 1 has the lowest score.

[0085] Step b2: Based on each parameter in the path parameters and the corresponding parameter weights, obtain the second path score for entering the lane after turning left for each path planning method.

[0086] Specifically, the parameter weights for detour parameters, turn failure parameters, left turn passage time, and direct access to the target lane are K1, K2, K3, and K4, respectively. K1, K2, K3, and K4 can be set based on requirements. When obtaining the second path score for each route planning method after a left turn, based on the parameters and their corresponding weights in the path parameters, it can be obtained by weighted summation of the parameters and their corresponding weights. That is, Y = K1*C1 + K2*C2 + K3*C3 + K4*C4. A higher second path score indicates that the route is the optimal route.

[0087] In this embodiment of the invention, if the length of the target lane is greater than the passage threshold, the second path score is given for directly turning left to the target lane and turning left from the current lane to another lane and then changing lanes to the target lane. The path of the vehicle turning left from the current lane to the target lane at the current intersection is planned, thereby improving the passage efficiency of the vehicle at the next intersection after turning left.

[0088] Compared to related technologies where left-turn control requires multiple lane changes to ensure the vehicle is in the correct lane, the left-turn path control method of this invention obtains the vehicle's target direction at the next intersection, determines the target lane for the left turn, and plans the path for the vehicle to turn left from the current lane to the target lane based on the relationship between the target lane length and the passage threshold. This allows the vehicle to obtain the target direction at the next intersection before turning left, thus determining the target lane for the vehicle when turning left at the current intersection and planning the path for the vehicle to turn left from the current lane to the target lane. This avoids incorrect target lane selection in the passable lanes, reduces the risk of left-turn failure, and improves overall traffic efficiency.

[0089] The following is a detailed explanation of the flow of the left-turn path control method provided in the embodiments of this application, using a specific example.

[0090] Reference Figure 4 Step 1: Obtain the target driving direction of the vehicle at the next intersection and determine the target lane corresponding to the left turn.

[0091] Specifically, the passable lanes after a vehicle turns left from its current lane at the intersection can be marked as Route ② and Route ④, while the passable lanes after other vehicles in the adjacent lane turn left can be marked as Route ① and Route ③.

[0092] Step 2: Obtain the lane length of the target lane.

[0093] Specifically, the length of the target lane between the current intersection and the next intersection is used as the lane length.

[0094] Step 3: Based on the relationship between the lane length of the target lane and the passage threshold, plan the path for the vehicle to turn left from the current lane at the current intersection to the target lane.

[0095] Specifically, if the length of the target lane is less than or equal to the passage threshold (D≤Dmin), it is necessary to determine whether the target lane is unique. If the target lane is unique, the vehicle directly enters the lane corresponding to the target driving direction. If the next intersection requires a right turn, the vehicle needs to enter the right-turn lane of the passable lane (target road), i.e., the lane corresponding to route ④. If the next intersection requires a left turn, the vehicle enters the left-turn lane of the passable lane in advance. If the next intersection requires going straight, the target lane is the straight lane. If the target lane is not unique, such as if the target driving direction is a right turn, and the right-turn lanes that can be entered after turning left from the current lane are numbered 3, 2, and 1 from right to left, then the first path from the current lane to the target lane numbered 3 has the highest score, the first path from the current lane to the target lane numbered 2 has the second highest score, and the first path from the current lane to the target lane numbered 1 has the lowest score. Therefore, the vehicle chooses the first path with the highest score to turn left to the target lane numbered 3. When planning the path for a vehicle to turn left from the current lane to the target lane using the control method of turning left directly from the current lane to the target lane, the vehicle's current position point and the center line of the target lane can be fitted, rather than fitting the target trajectory based on the current driving lane and the target lane. That is, when route ② exists in the target lane, the vehicle can drive over the lane line instead of only planning the trajectory according to the direction limited by the lane line, thereby avoiding the situation where the turning radius is too large.

[0096] Furthermore, if the length of the target lane is greater than the passage threshold (D > Dmin), then when determining the route planning method, including directly turning left to the target lane and turning left from the current lane to another lane and then changing lanes to the target lane, it is necessary to obtain the path parameters and corresponding parameter weights of the vehicle changing lanes from the current intersection to each lane. Based on each parameter and its corresponding weight in the path parameters, the second path score for entering the lane after turning left for each route planning method is obtained. The parameter weights for detour parameter C1, turn failure parameter C2, left turn passage time C3, and direct access parameter C4 to the target lane are K1, K2, K3, and K4, respectively. When obtaining the second path score for entering the lane after turning left for each route planning method based on each parameter and its corresponding weight, it can be obtained by weighted summation of each parameter and its corresponding weight. That is, Y = K1*C1 + K2*C2 + K3*C3 + K4*C4. A higher second path score indicates that the route is the optimal route.

[0097] In summary, the left-turn path control method of this invention combines congestion information and traffic threshold of the passable lane to select the target lane in the passable lane (target road), and then solves the optimal solution of the vehicle's target planned trajectory by adding path evaluation, thereby obtaining the optimal planned path of the vehicle. This solves and avoids the error of target lane selection in the passable lane (target road), reduces the risk of left turn failure, and improves the overall traffic efficiency.

[0098] Figure 5 A schematic diagram of an embodiment of the left-turn path control device of the present invention is shown. Figure 5 As shown, the device includes:

[0099] The lane acquisition module 510 is used to acquire the target driving direction of the vehicle at the next intersection and determine the target lane corresponding to the left turn.

[0100] The length acquisition module 520 is used to acquire the lane length of the target lane, which is the length of the target lane between the current intersection and the next intersection.

[0101] The path planning module 530 is used to plan the path for a vehicle to turn left from the current lane to the target lane at the current intersection based on the relationship between the lane length of the target lane and the passage threshold.

[0102] In one alternative implementation, the route planning module 530 includes:

[0103] The first threshold determination submodule is used to determine the route planning method as a direct left turn from the current lane to the target lane if the lane length of the target lane is less than or equal to the passage threshold, and to determine whether the target lane is unique.

[0104] The first path planning submodule is used to plan the path for the vehicle to turn left from the current lane to the target lane based on the path planning method if the target lane is unique.

[0105] The first path scoring submodule is used to obtain the first path score of the vehicle turning left from the current lane to each target lane if the target lane is not unique. The first path score is determined based on each target lane and the target driving direction.

[0106] The second path planning submodule is used to plan the path for the vehicle to turn left from the current lane to the target lane based on the path planning method and the first path score.

[0107] The second threshold determination submodule is used to determine the route planning method if the length of the target lane is greater than the passage threshold. This method includes turning left directly to the target lane and turning left from the current lane to another lane before changing lanes to the target lane.

[0108] The second path scoring submodule is used to obtain the second path score for entering the lane after turning left for each path planning method.

[0109] The third path planning submodule is used to plan the path for the vehicle to turn left from the current lane to the target lane at the current intersection based on the path planning method and the second path score.

[0110] In some optional implementations, the second path scoring submodule described above includes:

[0111] The parameter acquisition unit is used to acquire the path parameters and corresponding parameter weights of the vehicle changing lanes from the current intersection to each lane. The path parameters include the detour parameters of the vehicle turning left at the current intersection, the turning failure parameters, the left turn passage time, and the direct access parameters to the target lane.

[0112] The scoring calculation unit is used to obtain the second path score for entering the lane after turning left for each path planning method, based on each parameter in the path parameters and the corresponding parameter weights.

[0113] In some alternative implementations, the first path planning submodule described above includes:

[0114] The location acquisition unit is used to obtain the current location of the vehicle.

[0115] The position set unit is used to fit the current position of the vehicle to the center line of the target lane to obtain the fitting result.

[0116] The path planning unit is used to plan the path for the vehicle to turn left from the current lane to the target lane based on the fitting results.

[0117] In some alternative implementations, the first threshold determination submodule includes:

[0118] The congestion information acquisition unit is used to acquire the current congestion information of the passable lanes, which include the lanes between the current intersection and the next intersection.

[0119] The congestion level determination unit is used to determine the congestion level of passable lanes based on congestion information.

[0120] The passage threshold determination unit is used to determine the passage threshold based on the congestion level and the minimum lane change threshold, which is the minimum distance required for a vehicle to change lanes from the adjacent lane to the target lane after making a left turn.

[0121] The further functional descriptions of the above modules and units are the same as those in the corresponding method embodiments described above, and will not be repeated here.

[0122] The technical solution provided by the embodiments of the present invention, through the above-described device and its components, has the following advantages:

[0123] Compared to related technologies where left-turn control requires multiple lane changes to stay in the correct lane, the left-turn path control method of this invention obtains the target driving direction of the vehicle at the next intersection, determines the target lane for the left turn, and plans the path for the vehicle to turn left from the current lane to the target lane based on the relationship between the lane length of the target lane and the traffic threshold. This allows the vehicle to obtain the target driving direction of the next intersection in advance before turning left at the current intersection, thereby determining the target lane for the vehicle when turning left at the current intersection and planning the path for the vehicle to turn left from the current lane to the target lane, thus improving the traffic efficiency of the vehicle at the next intersection after turning left.

[0124] Figure 6 This diagram illustrates a structural schematic of an embodiment of a vehicle provided by the present invention. The specific embodiments of the present invention do not limit the specific implementation of the vehicle. The vehicle has the above-described... Figure 6 The left-turn path control device shown. The vehicle may include: a processor 602, a communications interface 604, a memory 606, and a communications bus 608.

[0125] The processor 602, communication interface 604, and memory 606 communicate with each other via communication bus 608. Communication interface 604 is used to communicate with other network elements such as clients or other servers. The processor 602 executes program 610, specifically performing the relevant steps described above in the method embodiment.

[0126] Specifically, program 610 may include program code, which includes computer-executable instructions.

[0127] Processor 602 may be a central processing unit (CPU), an application-specific integrated circuit (ASIC), or one or more integrated circuits configured to implement embodiments of the present invention. The vehicle may include one or more processors of the same type, such as one or more CPUs; or processors of different types, such as one or more CPUs and one or more ASICs.

[0128] Memory 606 is used to store program 610. Memory 606 may include high-speed RAM memory, and may also include non-volatile memory, such as at least one disk storage device.

[0129] Figure 7 Another structural schematic diagram of the vehicle provided by the present invention is shown. For example... Figure 7 As shown, the vehicle also includes: a rear corner millimeter-wave radar 2, a front millimeter-wave radar 3, a side-view camera 5, a front-view binocular camera 4, a lidar 8, a driver assistance function hard switch 1, a surround-view camera 6, an ultrasonic probe 7, a vehicle stability system 10, an electric power steering system 11, a vehicle controller 12, a body controller 13, an instrument cluster 14, a central control screen 15, and a remote monitoring module 16. The rear corner millimeter-wave radar 2, front millimeter-wave radar 3, side-view camera 5, front-view binocular camera 4, lidar 8, driver assistance function hard switch 1, surround-view camera 6, ultrasonic probe 7, vehicle stability system 10, electric power steering system 11, vehicle controller 12, body controller 13, instrument cluster 14, central control screen 15, and remote monitoring module 16 are all communicatively connected to the aforementioned processor 602. Furthermore, the remote monitoring module 16 can also establish communication connections with the cloud 18 and a mobile APP 17.

[0130] In this embodiment, the rear corner millimeter-wave radar 2 can be a 77GHz millimeter-wave radar, installed on the left and right sides inside the rear bumper, with a detection range of about 80m.

[0131] In this embodiment, the front millimeter-wave radar 3 can be a 77GHz millimeter-wave radar, installed at the front of the vehicle, with a detection range of about 160m.

[0132] In this embodiment, the side-view camera 5 can be a 100° wide-angle two-megapixel camera, with the side-front view arranged inside the rearview mirror and the side-rear view arranged above the fender, and the detection distance can reach about 70m.

[0133] In this embodiment, the forward-looking binocular camera 4 can be two cameras, with a field of view divided into small, medium and large angles, and the maximum detection distance can reach about 200m.

[0134] In this embodiment, the lidar 8 is installed at the junction of the vehicle roof and the windshield, and the maximum detection distance can reach about 250 meters.

[0135] The functions this vehicle can perform include adaptive cruise control, integrated cruise control, pilot assist, forward collision warning, automatic emergency braking, lane departure warning, lane keeping assist, lever lane change, and autonomous lane change.

[0136] like Figure 8As shown, the vehicle provided in this embodiment of the invention includes 3 millimeter-wave radars, 10 cameras, a processor 602, a vehicle stability system 10, an electric power steering system 11, a vehicle controller 12, a body controller 13, an instrument cluster 14, a central control screen 15, turn signals, etc. The sensor units communicate with the autonomous driving controller through a private CANFD network, and other related systems communicate with the autonomous driving controller through CANFD. The functions of the main related components are as follows:

[0137] The millimeter-wave radar is installed on the left and right sides of the rear bumper. It transmits radio waves (radar waves) and receives the echoes. The position data of the target is measured based on the time difference between transmission and reception. The detection range can reach 80m. Through millimeter waves, it can accurately detect the distance of obstacles from the vehicle and parameters such as relative speed.

[0138] The front millimeter-wave radar 3 is installed directly below the vehicle's license plate. It transmits radio waves (radar waves) and receives the echoes. It measures the target's position data based on the time difference between transmission and reception. The detection range can reach 160m. Through millimeter waves, it can accurately detect the distance and relative speed of obstacles from the vehicle.

[0139] The forward-facing binocular camera 4 is a combination of two high-pixel cameras with different viewing angles. It can detect obstacles up to about 200m away in front of it, identify lane line information, and identify vehicles entering and exiting at close range.

[0140] The side-view camera 5 can compensate for the poor recognition rate of the corner radar in low-speed scenarios. It can quickly and in advance capture the trend of other vehicles cutting in and close-range cutting in scenarios, so that the autonomous driving controller can process the cutting in and cutting out scenarios in advance.

[0141] The processor acquires information from the perception module (which includes millimeter-wave radar, a smart camera group, a side-view camera 5, and an integrated IMU) and uses algorithms to identify lane lines, vehicles on the road, curbs, obstacles, etc. It then rationally plans the trajectory for driver assistance and controls the vehicle's lateral and longitudinal directions. This enables functions such as following the vehicle in front when there are obstacles, maintaining a constant speed when there are no obstacles, avoiding rear-end collisions, stopping, and starting automatically. During the control process, the processor sends steering angle requests, deceleration requests, torque requests, etc., to various related systems.

[0142] The vehicle stability system 10 (ESC) is used to receive deceleration request commands sent by the autonomous driving controller, and at the same time feed back vehicle data such as deceleration, yaw angle, vehicle speed, and wheel speed for the processor to perform longitudinal control calculations.

[0143] Electric power steering (EPS) is used to execute steering angle and steering angle acceleration requests issued by the autonomous driving controller, controlling the steering wheel to turn to the angle commanded by the autonomous driving controller. If the EPS malfunctions or the driver intervenes to park, it needs to report the reason for disengaging from the autonomous driving controller.

[0144] The vehicle controller 12 (VCU) is used to receive torque requests from the autonomous driving controller, execute acceleration control, and provide real-time feedback on the vehicle's gear position and response torque.

[0145] The body control unit 13 (BCM) is used to receive control requests from the automatic driving control system for turn signals, hazard warning lights, windshield wipers, lights, etc.

[0146] Instrument cluster 14 (IC) is used to display the human-machine interface during the activation of driver assistance functions, providing text, images, and sound prompts.

[0147] The central control screen 15 (HU) displays the navigation assistance function during the activation process, showing the scene reconstruction interface and the user-customized settings entry.

[0148] The turn signals are used to respond to the illumination request of the body controller 13 during autonomous driving, alerting other vehicles to driving safety.

[0149] This invention also provides a computer-readable storage medium storing at least one executable instruction that, when executed on a vehicle / left-turn path control device, causes the vehicle / left-turn path control device to perform the left-turn path control method in any of the above method embodiments.

[0150] The algorithms or displays provided herein are not inherently related to any particular computer, virtual system, or other device. Furthermore, the embodiments of this invention are not directed to any particular programming language.

[0151] Numerous specific details are set forth in the specification provided herein. However, it will be understood that embodiments of the invention may be practiced without these specific details. Similarly, for the sake of brevity and to aid in understanding one or more aspects of the invention, in the description of exemplary embodiments of the invention above, various features of the embodiments are sometimes grouped together in a single embodiment, figure, or description thereof. The claims, which follow the detailed description, are hereby expressly incorporated into that detailed description, wherein each claim itself is a separate embodiment of the invention.

[0152] Those skilled in the art will understand that the modules in the device of the embodiment can be adaptively changed and placed in one or more devices different from that embodiment. Modules, units, or components in the embodiment can be combined into a single module, unit, or component, and further, they can be divided into multiple sub-modules, sub-units, or sub-components, except that at least some of such features and / or processes or units are mutually exclusive.

[0153] It should be noted that the above embodiments are illustrative of the invention and not restrictive, and that those skilled in the art can devise alternative embodiments without departing from the scope of the appended claims. In the claims, any reference signs placed between parentheses should not be construed as limiting the claims. The word "comprising" does not exclude the presence of elements or steps not listed in the claims. The word "a" or "an" preceding an element does not exclude the presence of a plurality of such elements. The invention can be implemented by means of hardware comprising several different elements and by means of a suitably programmed computer. In the unit claims enumerating several means, several of these means may be embodied by the same item of hardware. The use of the words first, second, and third, etc., does not indicate any order. These words can be interpreted as names. The steps in the above embodiments, unless otherwise specified, should not be construed as limiting the order of execution.

Claims

1. A left-turn path control method, characterized in that, Applied to bicycles, the method includes: Obtain the vehicle's target direction of travel at the next intersection and determine the target lane for a left turn; Obtain the lane length of the target lane, where the lane length is the length of the target lane between the current intersection and the next intersection; Obtain congestion information of the current road in the passable lanes, the passable lanes including the lanes between the current intersection and the next intersection; determine the congestion level of the passable lanes based on the congestion information; determine the passage threshold based on the congestion level and the minimum lane change threshold, the minimum lane change threshold being the minimum distance required for the vehicle to change lanes from the adjacent lane to the target lane after making a left turn; Based on the relationship between the lane length of the target lane and the passage threshold, the path for the vehicle to turn left from the current lane at the current intersection to the target lane is planned. If the length of the target lane is less than or equal to the passage threshold, and the target lane is not unique, then the first path score of the vehicle turning left from the current lane to each target lane is obtained; If the length of the target lane is greater than the passage threshold, the path parameters and corresponding parameter weights of the vehicle changing lanes from the current intersection to each lane are obtained. The path parameters include the detour parameters, turning failure parameters, left turn passage time, and direct access parameters to the target lane. Based on each parameter and corresponding parameter weight in the path parameters, a second path score is obtained for each path planning method corresponding to the left turn into the lane. Based on the second path score, the path of the vehicle turning left from the current lane of the current intersection to the target lane is planned. The detour parameter is determined based on the type of obstacle, the turn failure parameter is determined based on the turning radius before turning left into the passable lane, and the direct route parameter is determined based on the distance between the vehicle after turning left from the current lane into each target lane and the target driving direction at the next intersection.

2. The method according to claim 1, characterized in that, The method of planning the path for a vehicle to turn left from the current lane to the target lane at the current intersection based on the relationship between the lane length of the target lane and the traffic threshold includes: If the length of the target lane is less than or equal to the passage threshold, the path planning method is determined to be a direct left turn from the current lane to the target lane, and it is determined whether the target lane is unique. If the target lane is unique, then the path for the vehicle to turn left from the current lane to the target lane is planned based on the path planning method.

3. The method according to claim 2, characterized in that, The method of planning the path for a vehicle to turn left from the current lane to the target lane at the current intersection based on the relationship between the lane length of the target lane and the traffic threshold also includes: If the target lane is not unique, then obtain the first path score of the vehicle turning left from the current lane to each target lane. The first path score is determined based on each target lane and the target driving direction. Based on the path planning method and the first path score, the path for the vehicle to turn left from the current lane to the target lane is planned.

4. The method according to claim 2, characterized in that, The method of planning the path for a vehicle to turn left from the current lane to the target lane at the current intersection based on the relationship between the lane length of the target lane and the traffic threshold also includes: If the length of the target lane is greater than the passage threshold, then the path planning method is determined to include turning left directly to the target lane and turning left from the current lane to another lane and then changing lanes to the target lane. Obtain the second path score for the left turn lane corresponding to each of the path planning methods.

5. The method according to any one of claims 1 to 4, characterized in that, If the path planning method is to turn left directly from the current lane to the target lane, then the path for the vehicle to turn left from the current lane to the target lane is planned, including: Obtain the current position of the vehicle; The current position of the vehicle is fitted to the center line of the target lane to obtain the fitting result; Based on the fitting results, the path for the vehicle to turn left from the current lane to the target lane is planned.

6. A left-turn path control device, characterized in that, Applied to a bicycle, the device includes: The lane acquisition module is used to obtain the target driving direction of the vehicle at the next intersection and determine the target lane corresponding to the left turn; The length acquisition module is used to acquire the lane length of the target lane, wherein the lane length is the length of the target lane between the current intersection and the next intersection; The route planning module is used to obtain congestion information of the current road in the passable lanes, the passable lanes including the lanes between the current intersection and the next intersection; determine the congestion level of the passable lanes based on the congestion information; determine a passage threshold based on the congestion level and a minimum lane change threshold, the minimum lane change threshold being the minimum distance required for the vehicle to change lanes from the adjacent lane to the target lane after making a left turn; and plan the path for the vehicle to turn left from the current lane at the current intersection to the target lane based on the relationship between the lane length of the target lane and the passage threshold. The path planning module is specifically used to: if the length of the target lane is less than or equal to the passage threshold and the target lane is not unique, then obtain a first path score for the vehicle turning left from the current lane to each target lane; if the length of the target lane is greater than the passage threshold, then obtain the path parameters and corresponding parameter weights for the vehicle changing lanes from the current intersection to each lane. The path parameters include the detour parameters, turning failure parameters, left turn passage time, and direct access parameters to the target lane. Based on each parameter and corresponding parameter weight in the path parameters, a second path score is obtained for each path planning method corresponding to the left turn into the lane. Based on the second path score, the path for the vehicle to turn left from the current lane to the target lane at the current intersection is planned. The detour parameter is determined based on the type of obstacle, the turn failure parameter is determined based on the turning radius before turning left into the passable lane, and the direct route parameter is determined based on the distance between the vehicle after turning left from the current lane into each target lane and the target driving direction at the next intersection.

7. A vehicle, characterized in that, include: The processor, memory, communication interface, and communication bus are provided, wherein the processor, memory, and communication interface communicate with each other via the communication bus. The memory is used to store at least one executable instruction that causes the processor to perform the operation of the left-turn path control method as described in any one of claims 1-5.

8. A computer-readable storage medium, characterized in that, The storage medium stores at least one executable instruction, which, when executed on the vehicle / device, causes the vehicle / device to perform the operation of the left-turn path control method as described in any one of claims 1-5.

Images