Traveling method of vehicles on shunting road section

Abstract

The invention discloses a driving method of vehicles on a shunting road section, and relates to the technical field of intelligent driving. According to the navigation direction, the road condition of the candidate lane area and the navigation distance, a target candidate lane area is selected from candidate lane areas at the lane diversion position in front of the vehicle, and the driving track of the vehicle is determined according to a target lane area between the current lane area of the vehicle and the target candidate lane area, and the vehicle is controlled to run according to the track. According to the method, data of various sensors are not directly input into a model, and a selected path is output through the model, so that the difficulty of shunting and routing is relatively reduced; when the driving track is determined, the navigation direction, the road condition of the candidate lane area and the navigation distance are comprehensively considered, and the road condition of the candidate lane area can be multiple, namely, the driving track of the shunting road section is determined by comprehensively considering multiple factors, so that the shunting selected road is more accurate; the vehicle is controlled to select the appropriate road section to run in the shunting road section.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent driving, and particularly to a driving method for vehicles on a diversion section. Background Art

[0002] Diversion route selection in Navigate on Autopilot (NOA) is one of the core capabilities for intelligent driving vehicles to achieve autonomous navigation on complex urban roads, especially in scenarios such as multi-lane intersections and left / right turn intersections. The accuracy of diversion route selection directly determines the reliability, safety, and user experience of the system. An incorrect diversion route selection may cause the vehicle to miss a turn at an intersection, etc., and it is necessary to re-plan the route, increasing the driving distance.

[0003] In related diversion route selection schemes, the data of various sensors are directly input into the model, and the planned route is output through the model. However, the establishment of the model in this scheme is relatively difficult, and only some users have achieved the use of this method.

[0004] Therefore, how to reduce the difficulty of diversion route selection and control the vehicle to select a suitable section to drive on the diversion section is a technical problem that needs to be solved urgently by those skilled in the art. Summary of the Invention

[0005] The purpose of the present invention is to provide a driving method for vehicles on a diversion section to solve the problem of large difficulty in diversion route selection when directly inputting the data of various sensors into the model.

[0006] To solve the above technical problem, the present invention provides a driving method for vehicles on a diversion section, including:

[0007] Obtain lane information and a navigation direction; wherein, the lane information includes the lane information of the vehicle itself and the surrounding lane information of the lane where the vehicle itself is located; the lane information includes at least lane topological relationships;

[0008] Obtain a candidate lane area according to the lane information; wherein, the candidate lane area is the lane area at the diversion of the lane in front of the vehicle itself;

[0009] Obtain the road conditions and navigation distance of the candidate lane area; wherein, the road conditions of the candidate lane area include one or more of lane curvature, lane topological structure, traffic flow, and obstacles;

[0010] Select a target candidate lane area from all the candidate lane areas according to the navigation direction, the road conditions of the candidate lane area, and the navigation distance;

[0011] Obtain a target lane area between the current lane area where the vehicle is located and the target candidate lane area, determine a driving trajectory based on the target lane area, and control the vehicle to drive along the driving trajectory.

[0012] Exemplarily, the lane diversion is within a preset distance in front of the vehicle; determining the preset distance includes:

[0013] Obtain the first product result of the vehicle speed and a preset duration, and obtain the longitudinal distance;

[0014] Obtain the first result after adding the first product result and a first preset value, and obtain the second result after adding the longitudinal distance and a second preset value;

[0015] Obtain the minimum value between the first result and the second result;

[0016] Take the minimum value as the preset distance.

[0017] Exemplarily, determining the lane diversion includes:

[0018] Starting from the current lane area where the vehicle is located, traverse the lane areas in front of the vehicle according to the connection relationship of the lane topology;

[0019] Obtain the distance value between the current lane area where the vehicle is located and the currently traversed lane area;

[0020] When it is detected that the distance value is less than or equal to the preset distance, and it is detected that there are two or more successor lane areas in the currently traversed lane area, stop traversing the lane areas in front of the vehicle, and take the end point of the currently traversed lane area as the lane diversion;

[0021] When it is detected that the distance value is less than the preset distance, and it is detected that there is only one successor lane area in the currently traversed lane area, continue to traverse the lane areas in front of the vehicle, detect the relationship between the distance value and the preset distance, and detect the situation of the successor lanes of the currently traversed lane area. [[ID=3,2]]

[0022] Exemplarily, the selecting the target candidate lane area from all the candidate lane areas according to the navigation direction, the road conditions of the candidate lane area, and the navigation distance includes:

[0023] Determine the road condition cost of the candidate lane area and the navigation distance cost of the candidate lane area according to the navigation direction;

[0024] Determine the cost of the candidate lane area according to the road condition cost of the candidate lane area and the navigation distance cost of the candidate lane area;

[0025] Obtain the candidate lane area with the minimum cost from all the candidate lane areas;

[0026] Take the candidate lane area with the minimum cost as the target candidate lane area.

[0027] Exemplarily, the road conditions of the candidate lane area include lane curvature, lane topology, traffic flow, and obstacles; determining the cost of the candidate lane area according to the cost of the road conditions of the candidate lane area and the cost of the navigation distance of the candidate lane area includes:

[0028] Obtain the cost of the lane curvature of the candidate lane area, the cost of obtaining the lane topology, the cost of obtaining the traffic flow, the cost of obtaining the obstacles, and the cost of the navigation distance;

[0029] Determine the cost of the candidate lane area according to the sum of the cost of the lane curvature of the candidate lane area, the cost of obtaining the lane topology, the cost of obtaining the traffic flow, the cost of obtaining the obstacles, and the cost of the navigation distance.

[0030] Exemplarily, obtaining the cost of the lane curvature of the candidate lane area includes:

[0031] Collect the center points of the candidate lane area at a preset step size;

[0032] Perform denoising processing on the collected center points;

[0033] Determine the lane curvature of the candidate lane area according to the center points after denoising processing;

[0034] Obtain the second product result of the lane curvature and the gain coefficient of the curvature cost preset; wherein, the gain coefficient of the curvature cost is determined by the navigation direction, and when the navigation direction is straight, the gain coefficient of the curvature cost is larger;

[0035] Take the second product result as the cost of the lane curvature of the candidate lane area.

[0036] Exemplarily, obtaining the cost of the lane topology of the candidate lane area includes:

[0037] Obtain the consistency between the navigation direction and the splitting direction of the candidate lane area;

[0038] In the case of consistent directions, reduce the gain coefficient of the lane topology cost to determine the cost of the lane topology;

[0039] In the case of inconsistent directions, increase the gain coefficient of the lane topology cost or keep the gain coefficient of the lane topology cost unchanged to determine the cost of the lane topology of the candidate lane area.

[0040] Exemplarily, the cost of obtaining the traffic flow of the candidate lane area includes:

[0041] Obtaining the speeds of the obstacles on the candidate lane area, the deviation angles between the obstacles and the heading angle of the host vehicle, and obtaining the length of the candidate lane area;

[0042] Obtaining the third product result of the speed of each obstacle on the candidate lane area and the corresponding deviation angle;

[0043] Based on the first ratio of the sum of the third product results of all obstacles to the length of the candidate lane area, determining the dynamic density index of the candidate lane area;

[0044] Obtaining the gain coefficient of the traffic flow cost;

[0045] According to the second ratio of the gain coefficient of the traffic flow cost to the dynamic density index of the candidate lane, taking the second ratio as the cost of the traffic flow of the candidate lane area.

[0046] Exemplarily, the cost of obtaining the obstacles in the candidate lane area includes:

[0047] Obtaining the third ratio of the area of each obstacle to the area of the candidate lane area;

[0048] Obtaining the fourth product result of the sum of the third ratios corresponding to all obstacles and the obstacle gain coefficient; wherein, the obstacle gain coefficient has a positive correlation with the number of obstacles on the candidate lane area;

[0049] Taking the fourth product result as the cost of the obstacles in the candidate lane area.

[0050] Exemplarily, the cost of obtaining the navigation distance of the candidate lane area includes:

[0051] Obtaining the minimum distance value between the navigation distance of the candidate lane area and the distance threshold;

[0052] Obtaining the fifth product result of the minimum distance value and the navigation distance gain coefficient; wherein, the navigation distance gain coefficient when the navigation direction is not straight is greater than the navigation distance gain coefficient when the navigation direction is straight;

[0053] Taking the negative value of the fifth product result as the cost of the navigation distance of the candidate lane area.

[0054] In the driving method of vehicles on a diversion section provided by the present invention, first, a candidate lane area at the diversion of the lane in front of the vehicle is determined according to lane information. After obtaining the navigation direction, the road conditions of the candidate lane area, and the navigation distance, a target candidate lane area is selected from all the candidate lane areas according to the navigation direction, the road conditions of the candidate lane area, and the navigation distance. Finally, the driving trajectory of the vehicle is determined according to the target lane area between the current lane area of the vehicle and the target candidate lane area, and then the vehicle is controlled to drive along the determined trajectory. In this method, the data of various sensors are not directly input into the model, and the selected path is output through the model. Therefore, the difficulty of diversion route selection is relatively reduced; when determining the driving trajectory, the navigation direction, the road conditions of the candidate lane area, and the navigation distance are comprehensively considered, and the road conditions of the candidate lane area may include multiple items such as lane curvature, lane topology, traffic flow, and obstacles, that is, the driving trajectory of the diversion section is determined by comprehensively considering various factors, making the selected diversion route more accurate, and realizing the control of the vehicle to select a suitable section to drive on the diversion section. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] To more clearly illustrate the embodiments of the present invention, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0056] Figure 1 Schematic diagram of an intelligent driving system provided by an embodiment of the present invention;

[0057] Figure 2 Schematic diagram of the first main road diversion scenario provided by an embodiment of the present invention;

[0058] Figure 3 Schematic diagram of the second main road diversion scenario provided by an embodiment of the present invention;

[0059] Figure 4 Schematic diagram of a diversion scenario before a right-turn intersection provided by an embodiment of the present invention;

[0060] Figure 5 Schematic diagram of a diversion scenario before a left-turn intersection provided by an embodiment of the present invention;

[0061] Figure 6 Flowchart of a driving method of vehicles on a diversion section provided by an embodiment of the present invention;

[0062] Figure 7 is Figure 4 Schematic diagram of the lane topological relationship of the scenario shown;

[0063] Figure 8 Structural diagram of the electronic device provided by an embodiment of the present invention. Detailed implementation manners

[0064] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the protection scope of the present invention.

[0065] The core of the present invention is to provide a driving method for vehicles on a diversion section to solve the problem of great difficulty in diversion route selection when directly inputting data of various sensors into a model.

[0066] Figure 1 Schematic diagram of an intelligent driving system provided by an embodiment of the present invention, as Figure 1 shown, the system includes: a perception function module, a map function module, a navigation function module, a decision-making and planning function module, and a control function module.

[0067] The perception function module processes real-time visual information and outputs lane information such as lane center line and boundary line, obstacle information, traffic light information, etc.; the map function module receives perception information to construct a static map and outputs it together with navigation information and map data; the decision-making and planning function module receives perception information and map navigation information and outputs a smooth driving trajectory; the control function module mainly receives the driving trajectory output by the decision-making and planning module and calculates the lateral steering wheel angle and longitudinal acceleration to control the vehicle. The process of diversion route selection provided by the present invention is implemented in the decision-making and planning function module as the core solution.

[0068] This solution is suitable for various diversion scenarios such as main road diversion and diversion before left and right turn intersections. Figure 2 Schematic diagram of the first main road diversion scenario provided by an embodiment of the present invention. Figure 2 In it, two lanes are evenly diverted from each of the following lanes. Figure 3 Schematic diagram of the second main road diversion scenario provided by an embodiment of the present invention. Figure 3 In it, two lanes are diverted from one lane below. Figure 4 Schematic diagram of a diversion scenario before a right turn intersection provided by an embodiment of the present invention. Figure 4 In it, two lanes are diverted from the rightmost lane below. Figure 5 Schematic diagram of a diversion scenario before a left turn intersection provided by an embodiment of the present invention. Figure 5 In it, two lanes are diverted from the leftmost lane below.

[0069] In order to be able to select a suitable road for driving in a diversion scenario, an embodiment of the present invention provides a driving method for vehicles on a diversion section. Figures 2 to 5 The sections in the diversion scenario described in Figures 2 to 5 are all diversion sections, and a suitable lane is selected for the vehicle to drive on the diversion section.

[0070] In order to enable those skilled in the art to better understand the solution of the present invention, the present invention will be further described in detail below with reference to the drawings and specific embodiments. Figure 6 It is a flowchart of a driving method for vehicles on a diversion section provided by an embodiment of the present invention. As Figure 6 shown, the method includes:

[0071] S10: Obtain lane information and navigation direction;

[0072] Obtain lane information and navigation information from the map function module. The lane information includes the lane information where the vehicle is located and the surrounding lane information of the lane where the vehicle is located. Such as the lane where the vehicle is located, lane topological relationship. The navigation information is such as a list of intersection turning information: going straight or turning left, right, or making a U-turn, etc. Figure 7 For Figure 4 shown in the schematic diagram of the lane topological relationship of the scenario, as Figure 7 shown, it includes lane1 to lane10.

[0073] S11: Obtain a candidate lane area according to the lane information.

[0074] The candidate lane area is the lane area (referred to as lane) at the diversion of the lane in front of the vehicle. Figure 7 In Figure 7 , the vehicle is located below the last lane and travels upward (i.e., in the direction of the front of the vehicle). Figure 7 In Figure 7 , the lane diversion is the end position of lane4, that is, the starting positions of lane7 and lane8.

[0075] In order to determine the candidate lane area, it is first necessary to find the lane diversion in front of the vehicle. In order to make the obtained driving trajectory more accurate, select the lane diversion at a preset distance in front of the vehicle. That is, the lane diversion is within a preset distance in front of the vehicle. The value of the preset distance is not limited and is determined according to the actual situation. The preset distance is at least determined by the vehicle speed of the vehicle, a preset duration, and the longitudinal distance (the driving direction of the vehicle along the lane) between the lane diversion and the vehicle. The preset duration is not limited and is calibrated according to the actual situation.

[0076] If the preset distance is only determined according to the vehicle speed of the vehicle, the preset duration, and the longitudinal distance between the lane diversion and the vehicle, considering the case where the vehicle speed is 0 and the longitudinal distance is 0, then the distance between the diversion and the vehicle is 0, that is, it is impossible to search in front of the vehicle to determine whether there is a lane diversion. Therefore, a first preset value and a second preset value are set here.

[0077] Determining a preset distance includes:

[0078] Obtaining a first product result of the vehicle speed of the host vehicle and a preset duration, and obtaining a longitudinal distance;

[0079] Obtaining a first result after adding the first product result and a first preset value, and obtaining a second result after adding the longitudinal distance and a second preset value;

[0080] Obtaining the minimum value between the first result and the second result;

[0081] Taking the minimum value as the preset distance.

[0082] The lane diversion point is within the preset distance in front of the host vehicle. In order to determine the lane diversion point, in implementation, determining the lane diversion point includes:

[0083] Starting from the lane area where the host vehicle is currently located, traversing the lane area in front of the host vehicle according to the connection relationship of the lane topology;

[0084] Obtaining a distance value between the lane area where the host vehicle is currently located and the currently traversed lane area;

[0085] When it is detected that the distance value is less than or equal to the preset distance, and it is detected that there are two or more successor lane areas in the currently traversed lane area, stop traversing the lane area in front of the host vehicle, and take the end point of the currently traversed lane area as the lane diversion point;

[0086] When it is detected that the distance value is less than the preset distance, and it is detected that there is only one successor lane area in the currently traversed lane area, continue to traverse the lane area in front of the host vehicle, detect the relationship between the distance value and the preset distance, and detect the situation of the successor lanes of the currently traversed lane area.

[0087] Figure 7 In, when the currently traversed lane area is lane1, its successor lane is lane4; when the currently traversed lane area is lane4, its successor lanes are lane7 and lane8.

[0088] In addition, when it is detected that the distance value is greater than the preset distance, stop traversing the lane area in front of the host vehicle, and end.

[0089] To enable those skilled in the art to better understand the above determination of the lane diversion point and the determination of the candidate lane area, the following still takes Figure 7 as an example to describe this process in detail.

[0090] First, obtain vehicle-related information such as the lane where the vehicle is located (e.g., lane1), the lane topological relationship, and navigation information such as the list of intersection turning information (e.g., going straight, turning left, turning right, or making a U-turn) from the map function module. Based on the above information, perform a lane topology traversal to determine whether there is a diversion within a certain threshold in front of the vehicle (i.e., the preset distance, and a dynamic threshold can be set according to the vehicle speed), and obtain the forward driving direction based on the navigation information.

[0091] Specifically, first determine whether there is a diversion within a certain threshold in front of the vehicle. Obtain the lane area where the vehicle is located (e.g., lane1), the lane information around the vehicle, and the intersection information from the map function module. Since it is not necessary to consider the road structure after a too distant section or intersection, a distance threshold can be set based on the vehicle speed and the longitudinal distance between the intersection and the vehicle. , set a distance threshold ;

[0092] ;

[0093] Among them is the real-time vehicle speed, t is a calibratable time parameter (i.e., the preset duration), is the calibratable buffer margin, α is the first preset value, and β is the second preset value. Setting α is to prevent non-forward traversal when the speed is 0. The value of α can be set by multiplying the urban road speed limit by 1s, such as 40m or 50m. Setting β is to prevent traversing into or after the intersection. A relatively small value can be set for β, such as 1m.

[0094] Starting from the lane where the vehicle is located as the search starting point, traverse along the connection relationship of the lane topology and accumulate the geometric length of the lane. When the accumulated length exceeds the threshold , terminate the search. If it is detected that there are two or more successor lane areas in a certain lane area during the traversal process, terminate the search. Store all the successor lane areas at the diversion point in the candidate set, calculate the cost of each successor lane area, and finally select a lane with the lowest cost.

[0095] The following Figure 7 is an example. Starting from the vehicle lane lane1, traverse to find the successor lane area. The initial accumulated lane length , traverse to the successor lane area lane4, and accumulate the lane length , is less than the threshold , continue to search for the successor lane area. When it is found that there are two successor lane areas lane7 and lane8 in lane4, terminate the traversal, thereby determining that there is a diversion in front of the vehicle, and store these two lane areas in the candidate set.

[0096] S12: Obtain the road conditions and navigation distance of the candidate lane area;

[0097] S13: Select a target candidate lane area from all candidate lane areas according to the navigation direction, the road conditions of the candidate lane area, and the navigation distance.

[0098] To determine the target candidate lane area, it is necessary to obtain the road conditions of the candidate lane area, as well as the navigation distance and navigation direction. It should be noted that according to the above method, multiple lane diverging points may be determined. In order to accurately divert and select a route, in practice, select the lane diverging point closest to the vehicle itself. Determine the driving direction of the intersection closest to the vehicle itself. Obtain navigation information from the map function module: The intersection turning information list includes the intersection turning direction and distance. According to the validity of the navigation and the distance of the intersection, obtain the first valid turning direction closest to the vehicle itself. 。

[0099] In the present invention, the road conditions include not only lane information (such as lane topology), but also traffic and obstacles on the road, etc. In order to make the determined target candidate lane area more accurate, the road conditions of the candidate lane area include one or more of lane curvature, lane topology, traffic flow, and obstacles.

[0100] Selecting a target candidate lane area from all candidate lane areas according to the navigation direction, the road conditions of the candidate lane area, and the navigation distance includes:

[0101] Determine the cost of the road conditions of the candidate lane area and the cost of the navigation distance of the candidate lane area according to the navigation direction;

[0102] Determine the cost of the candidate lane area according to the cost of the road conditions of the candidate lane area and the cost of the navigation distance of the candidate lane area;

[0103] Obtain the candidate lane area with the minimum cost from all candidate lane areas;

[0104] Take the candidate lane area with the minimum cost as the target candidate lane area.

[0105] In practice, when the road conditions of the candidate lane area include lane curvature, lane topology, traffic flow, and obstacles; determining the cost of the candidate lane area according to the cost of the road conditions of the candidate lane area and the cost of the navigation distance of the candidate lane area includes:

[0106] Obtain the cost of the lane curvature of the candidate lane area, the cost of obtaining the lane topology, the cost of obtaining the traffic flow, the cost of obtaining the obstacles, and the cost of the navigation distance;

[0107] Determine the cost of the candidate lane area based on the sum of the cost of the lane curvature of the candidate lane area, the cost of obtaining the lane topology, the cost of obtaining the traffic flow, the cost of obtaining obstacles, and the cost of the navigation distance.

[0108] The following specifically describes the determination process of each cost.

[0109] 1. Cost of lane curvature:

[0110] The cost of obtaining the lane curvature of the candidate lane area includes:

[0111] Collect the center points of the candidate lane area at a preset step size;

[0112] Denoise the collected center points;

[0113] Determine the lane curvature of the candidate lane area based on the denoised center points;

[0114] Obtain the second product result of the lane curvature and the gain coefficient of the preset curvature cost; wherein, the gain coefficient of the curvature cost is determined by the navigation direction, and when the navigation direction is straight, the greater the gain coefficient of the curvature cost;

[0115] Take the second product result as the cost of the lane curvature of the candidate lane area.

[0116] There is no limit to the preset step size, which is determined according to the actual situation. When determining the lane curvature of the candidate lane area based on the denoised center points, the three-point method can be used to calculate the curvature or the curvature can be determined by polynomial fitting.

[0117] Obtain the center points of the candidate lane area from the map function module and calculate the average curvature using the three-point method. First, sample the center points of the candidate lane area. To control the granularity of curvature calculation, the present technology dynamically adjusts the sampling step size according to the vehicle speed , increasing the step size at high speed and decreasing the step size at low speed, with the range controlled between 1m and 5m. Starting from the first point, sample one point every meter until no center point can be obtained. To improve the stability of curvature calculation, the present technology uses the median filtering method to denoise the sampled points. This method is a common data denoising method and will not be described in the present technology. Loop through the denoised sampled points, take out 3 adjacent points each time, calculate the curvature of these 3 points through the following formula, set the cumulative curvature quantity as U and the curvature sum, and finally calculate the average curvature of the sampled points as the curvature of the candidate lane. The calculation formula of the three-point method is as follows:

[0118] ;

[0119] where a, b, and c are the three side lengths of the triangle formed by the 3 points respectively, and p is the semi-perimeter of the triangle.

[0120] Cost of lane curvature The calculation formula is as follows:

[0121] ;

[0122] Where is the gain coefficient of the curvature cost. When the navigation driving direction is straight ahead, this coefficient can be significantly increased, making the curvature cost term the dominant factor. is the curvature of the lane, is the curvature calculated using the three-point method.

[0123] 2. Cost of lane topology:

[0124] The cost of obtaining the lane topology of the candidate lane area includes:

[0125] Obtaining the consistency between the navigation direction and the diversion direction of the candidate lane area;

[0126] In the case of consistent directions, reducing the gain coefficient of the lane topology cost to determine the cost of the lane topology;

[0127] In the case of inconsistent directions, increasing the gain coefficient of the lane topology cost or keeping the gain coefficient of the lane topology cost unchanged to determine the lane topology cost of the candidate lane area.

[0128] Obtaining the diversion attributes of the candidate lane area from the map function module, taking Figure 7 as an example, the attribute of lane7 is right diversion, and the attribute of lane8 is being right-diverted. If the navigation driving direction is the same as the diversion direction of the candidate lane area, the total cost is affected through a negative cost reward mechanism. For example, if Figure 7 the navigation driving direction is a right turn, the cost of the lane area lane7 with right diversion is reduced. The cost of the lane topology The calculation formula is as follows:

[0129] ;

[0130] Where is the gain coefficient of the lane topology cost, which can be adjusted according to the driving direction and different scenarios to control the contribution of the lane topology to the total cost.

[0131] 3. Cost of traffic flow:

[0132] The cost of obtaining the traffic flow of the candidate lane area includes:

[0133] Obtaining the speeds of various obstacles on the candidate lane area, the deviation angles between each obstacle and the heading angle of the vehicle itself, and obtaining the length of the candidate lane area;

[0134] Obtain the third product result of the speed of each obstacle on the candidate lane area and the corresponding deviation angle;

[0135] Based on the first ratio of the sum of the third product results of all obstacles to the length of the candidate lane area, determine the dynamic density index of the candidate lane area;

[0136] Obtain the gain coefficient of the traffic flow cost;

[0137] According to the second ratio of the gain coefficient of the traffic flow cost to the dynamic density index of the candidate lane, use the second ratio as the cost of the traffic flow in the candidate lane area.

[0138] To comprehensively evaluate the dynamic characteristics of the traffic flow, this technology uses the dynamic density index to reflect the impact of the traffic flow on the lane cost, comprehensively considering vehicle speed, direction and lane spatial distribution. Obtain the data of dynamic obstacles from the perception function module, including position, heading and speed, and obtain the lane length of the candidate lane from the map function module. First, filter the obstacles, only need to consider the obstacles within a certain range in front of the vehicle itself, filter out the vehicles behind the vehicle itself and those at a relatively long distance, and filter out the vehicles located outside the candidate lane according to the position of the obstacles. The cost of the traffic flow The calculation formula is as follows:

[0139] ;

[0140] Where is the gain coefficient of the traffic flow cost, is the dynamic density index of the candidate lane, which is inversely proportional to the cost, is the vehicle speed of the i-th obstacle, is the number of obstacles, is the deviation from the vehicle's heading angle, is the lane length of the candidate lane.

[0141] 4. Cost of obstacles (static obstacles):

[0142] Obtain the third ratio of the area of each obstacle to the area of the candidate lane area;

[0143] Obtain the fourth product result of the sum of the third ratios corresponding to all obstacles and the obstacle gain coefficient; among them, the obstacle gain coefficient is positively correlated with the number of obstacles on the candidate lane area;

[0144] Use the fourth product result as the cost of the obstacles in the candidate lane area.

[0145] In the split routing, static obstacles such as construction areas, illegally parked vehicles, road closures, etc. will significantly affect the feasibility and safety of lanes. This technology quantifies the impact of static obstacles on lane costs through the occupancy rate of static obstacles. The data of static obstacles obtained from the perception function module includes the position, length, and width. The lane length and width of the candidate lane are obtained from the map function module, and the cost of the static obstacle occupancy rate is calculated. The cost of the obstacle is calculated as follows:

[0146] ;

[0147] where is the occupancy rate of the static obstacle, are the length and width of the th static obstacle respectively, is the number of static obstacles, are the length and width of the candidate lane area respectively, is the static obstacle gain coefficient, which controls the contribution weight of the static obstacle to the total cost and can be adjusted according to different driving scenarios. For example Figure 7 in, if the driving direction is right turn, if there is a static obstacle in the split lane area lane7, the coefficient can be increased, and it is preferred to select the left lane area Lane8.

[0148] 5. Cost of navigation distance:

[0149] The cost of obtaining the navigation distance of the candidate lane area includes:

[0150] Obtaining the minimum distance value between the navigation distance of the candidate lane area and the distance threshold;

[0151] Obtaining the fifth product result of the minimum distance value and the navigation distance gain coefficient; among them, the navigation distance gain coefficient when the navigation direction is not straight is greater than the navigation distance gain coefficient when the navigation direction is straight;

[0152] Taking the negative value of the fifth product result as the cost of the navigation distance of the candidate lane area.

[0153] The navigation distance refers to the remaining drivable length of the current candidate lane area in the global planned path, usually the cumulative distance from the starting point of the lane area to the end point of the navigation path. In split routing, lanes with longer navigation distances are preferred to ensure the consistency of the local path and the global plan. The navigation distance affects the total cost through a negative cost reward mechanism. The cost of the navigation distance is calculated as follows:

[0154] ;

[0155] where is the navigation distance gain coefficient, which controls the contribution weight of the navigation distance to the total cost. When the navigation driving direction is not straight ahead, this coefficient can be appropriately increased to avoid missing the navigation. is the navigation distance of the candidate lane, is the calibratable time parameter, is the maximum effective distance threshold to avoid overly long lanes from dominating the cost calculation.

[0156] Calculate the total cost of each candidate lane area by combining the above 5 factors The calculation formula is as follows:

[0157] ;

[0158] When it is determined that there is a diversion in front of the vehicle itself, calculate the total cost for each candidate lane area according to the driving direction and combine the above factors, and regard the lane area with the lowest cost as the lane with the highest priority.

[0159] S14: Obtain the target lane area between the current lane area of the vehicle itself and the target candidate lane area, determine the driving trajectory according to the target lane area, and control the vehicle to drive along the driving trajectory.

[0160] Longitudinally connect the self-lane area and the successor lane area with the highest priority to form a driving channel, traverse the lane areas in the driving channel, connect the center lines to form a reference line as the initial rough trajectory, and subsequent lateral optimization and longitudinal speed planning can be performed based on this rough trajectory to generate a smooth trajectory for the control function module. Take Figure 7 as an example. The vehicle is in lane1, and the candidate lane areas at the diversion are lane8 and lane7. Obtain the target candidate lane area from the costs of lane8 and lane7, specifically obtain the costs of lane8 and lane7, and select the target lane according to the costs. If lane7 is selected as the target candidate lane area, connect lane1, lane4, and lane7 to form a trajectory line, and control the vehicle to drive along this trajectory line.

[0161] In the above embodiment, a driving method for a vehicle in a diversion section is described in detail. The present invention also provides an embodiment corresponding to a driving device for a vehicle in a diversion section. It should be noted that the present invention describes the embodiments of the device part from two perspectives, one is from the perspective of functional modules, and the other is from the perspective of hardware.

[0162] An embodiment of the present invention provides a driving control device for a vehicle in a diversion section. From the perspective of functional modules, it includes:

[0163] The first acquisition module is configured to acquire lane information and a navigation direction; wherein, the lane information includes the lane information of the vehicle itself and the surrounding lane information of the lane where the vehicle itself is located; the lane information includes at least a lane topological relationship;

[0164] The second acquisition module is configured to acquire a candidate lane area according to the lane information; wherein, the candidate lane area is the lane area at the lane diversion ahead of the vehicle itself;

[0165] The third acquisition module is configured to acquire the road conditions and the navigation distance of the candidate lane area; wherein, the road conditions of the candidate lane area include one or more of lane curvature, lane topological structure, traffic flow, and obstacles;

[0166] The selection module is configured to select a target candidate lane area from all the candidate lane areas according to the navigation direction, the road conditions of the candidate lane area, and the navigation distance;

[0167] The acquisition and control module is configured to acquire a target lane area between the lane area where the vehicle itself is currently located and the target candidate lane area, determine a driving trajectory according to the target lane area, and control the vehicle itself to drive along the driving trajectory.

[0168] Since the embodiments of the device part correspond to the embodiments of the method part, for the embodiments of the device part, please refer to the description of the embodiments of the method part, which will not be elaborated here for the time being.

[0169] This embodiment is from a hardware perspective. As Figure 8 shown, the electronic device includes:

[0170] A memory 20, configured to store a computer program;

[0171] A processor 21, configured to implement the steps of the method for the driving of a vehicle in a diversion section as mentioned in the above embodiments when executing the computer program.

[0172] Among them, the processor 21 may include one or more processing cores, such as a 4-core processor, an 8-core processor, etc. The processor 21 may be implemented in at least one hardware form of a Digital Signal Processor (DSP), a Field-Programmable Gate Array (FPGA), or a Programmable Logic Array (PLA). The processor 21 may also include a main processor and a coprocessor. The main processor is a processor used to process data in the wake state, also known as the Central Processing Unit (CPU); the coprocessor is a low-power processor used to process data in the standby state. In some embodiments, the processor 21 may be integrated with a Graphics Processing Unit (GPU), and the GPU is responsible for rendering and drawing the content to be displayed on the display screen. In some embodiments, the processor 21 may also include an Artificial Intelligence (AI) processor, which is used to process computing operations related to machine learning.

[0173] The memory 20 may include one or more computer-readable storage media, and the computer-readable storage media may be non-transitory. The memory 20 may also include high-speed random access memory and non-volatile memory, such as one or more disk storage devices and flash storage devices. In this embodiment, the memory 20 is at least used to store the following computer program 201. After the computer program is loaded and executed by the processor 21, it can implement the relevant steps of the driving method of the vehicle on the shunt section disclosed in any of the foregoing embodiments. In addition, the resources stored in the memory 20 may also include an operating system 202 and data 203, etc., and the storage method may be temporary storage or permanent storage. Among them, the operating system 202 may include Windows, Unix, Linux, etc. The data 203 may include, but is not limited to, the data involved in the driving method of the vehicle on the shunt section mentioned above.

[0174] In some embodiments, the electronic device may further include a display screen 22, an input / output interface 23, a communication interface 24, a power supply 25, and a communication bus 26.

[0175] Those skilled in the art can understand that Figure 8 the structure shown in

[0176] The electronic device provided by the embodiment of the present invention includes a memory and a processor. When the processor executes the program stored in the memory, the following method can be implemented: the driving method of vehicles on the shunt section, with the same effect as above.

[0177] Finally, the present invention also provides an embodiment corresponding to a computer-readable storage medium. A computer program is stored on the computer-readable storage medium, and when the computer program is executed by the processor, the steps recorded in the above method embodiments are implemented.

[0178] It can be understood that if the method in the above embodiments is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and executes all or part of the steps of the methods described in the various embodiments of the present invention. The foregoing storage media include: USB flash drives, mobile hard disks, read-only memories (ROMs), random access memories (RAMs), magnetic disks, or optical discs, etc., which can store program codes.

[0179] The computer-readable storage medium provided by the present invention includes the above-mentioned driving method of vehicles on the shunt section, with the same effect as above.

[0180] The above has introduced in detail a driving method of vehicles on a shunt section provided by the present invention. The various embodiments in the specification are described in a progressive manner. Each embodiment focuses on the differences from other embodiments. The same or similar parts among the various embodiments can be referred to each other. For the device disclosed in the embodiments, since it corresponds to the method disclosed in the embodiments, the description is relatively simple, and the relevant parts can be referred to the description of the method part. It should be noted that for those of ordinary skill in the art in this technical field, without departing from the principle of the present invention, several improvements and modifications can be made to the present invention, and these improvements and modifications also fall within the protection scope of the present invention.

[0181] It should also be noted that in this specification, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the term "comprising", "including" or any other variant thereof is intended to cover non-exclusive inclusion, such that a process, method, article or device comprising a series of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article or device. Without further limitation, an element defined by the statement "comprising an..." does not exclude the presence of additional identical elements in the process, method, article or device comprising the element.

Claims

1. A driving method for vehicles on a shunt section, characterized in that Including: Obtain lane information and navigation direction; wherein, the lane information includes the lane information of the vehicle itself and the surrounding lane information of the lane where the vehicle itself is located; the lane information includes at least lane topological relationships; Obtain a candidate lane area according to the lane information; wherein, the candidate lane area is the lane area at the lane diversion ahead of the vehicle itself; Obtain the road conditions and navigation distance of the candidate lane area; wherein, the road conditions of the candidate lane area include one or more of lane curvature, lane topological structure, traffic flow, and obstacles; Select a target candidate lane area from all the candidate lane areas according to the navigation direction, the road conditions of the candidate lane area, and the navigation distance; Obtain the target lane area between the current lane area where the vehicle itself is located and the target candidate lane area, determine the driving trajectory according to the target lane area, and control the vehicle itself to drive along the driving trajectory.

2. The driving method of the vehicle on the shunt section according to claim 1, wherein, The lane diversion is within a preset distance ahead of the vehicle itself; Determining the preset distance includes: Obtain the first product result of the vehicle speed of the vehicle itself and a preset duration, and obtain the longitudinal distance; Obtain the first result after adding the first product result and a first preset value, and obtain the second result after adding the longitudinal distance and a second preset value; Obtain the minimum value between the first result and the second result; Take the minimum value as the preset distance.

3. The driving method of the vehicle on the shunt section according to claim 2, characterized in that Determining the lane diversion includes: Starting from the current lane area where the vehicle itself is located, traverse the lane area ahead of the vehicle itself according to the connection relationship of the lane topology; Obtain the distance value between the current lane area where the vehicle itself is located and the currently traversed lane area; When it is detected that the distance value is less than or equal to the preset distance, and it is detected that there are two or more successor lane areas in the currently traversed lane area, stop traversing the lane area ahead of the vehicle itself, and take the end point of the currently traversed lane area as the lane diversion; When it is detected that the distance value is less than the preset distance, and it is detected that there is only one successor lane area in the currently traversed lane area, continue to traverse the lane area ahead of the vehicle itself, detect the relationship between the distance value and the preset distance, and detect the situation of the successor lane of the currently traversed lane area.

4. The driving method of a vehicle on a shunt section according to any one of claims 1 to 3, characterized in that The selecting a target candidate lane area from all the candidate lane areas according to the navigation direction, the road conditions of the candidate lane area, and the navigation distance includes: Determine the cost of the road conditions of the candidate lane area and the cost of the navigation distance of the candidate lane area according to the navigation direction; Determine the cost of the candidate lane area according to the cost of the road conditions of the candidate lane area and the cost of the navigation distance of the candidate lane area; Obtain the candidate lane area with the minimum cost from all the candidate lane areas; Take the candidate lane area with the minimum cost as the target candidate lane area.

5. The driving method of the vehicle in the shunt section according to claim 4, characterized in that The road conditions of the candidate lane area include lane curvature, lane topological structure, traffic flow, and obstacles; the determining the cost of the candidate lane area according to the cost of the road conditions of the candidate lane area and the cost of the navigation distance of the candidate lane area includes: The cost of obtaining the lane curvature of the candidate lane region, the cost of obtaining the lane topology, the cost of obtaining the traffic flow, the cost of obtaining the obstacles, and the cost of the navigation distance; Determine the cost of the candidate lane region based on the sum of the cost of the lane curvature of the candidate lane region, the cost of obtaining the lane topology, the cost of obtaining the traffic flow, the cost of obtaining the obstacles, and the navigation distance.

6. The driving method of a vehicle on a shunt section according to claim 5, characterized in that, The cost of obtaining the lane curvature of the candidate lane region includes: Collect the center points of the candidate lane region at a preset step size; Perform denoising processing on the collected center points; Determine the lane curvature of the candidate lane region based on the denoised center points; Obtain the second product result of the lane curvature and the gain coefficient of the preset curvature cost; wherein, the gain coefficient of the curvature cost is determined by the navigation direction, and when the navigation direction is straight, the greater the gain coefficient of the curvature cost; Take the second product result as the cost of the lane curvature of the candidate lane region.

7. The driving method of the vehicle on the shunt section according to claim 5, wherein, The cost of obtaining the lane topology of the candidate lane region includes: Obtain the consistency between the navigation direction and the diverging direction of the candidate lane region; In the case of consistent directions, reduce the gain coefficient of the lane topology cost to determine the cost of the lane topology; In the case of inconsistent directions, increase the gain coefficient of the lane topology cost or keep the gain coefficient of the lane topology cost unchanged to determine the cost of the lane topology of the candidate lane region.

8. The driving method of the vehicle on the shunt section according to claim 5, characterized in that The cost of obtaining the traffic flow of the candidate lane region includes: Obtain the speed of each obstacle on the candidate lane region, the deviation angle between each obstacle and the heading angle of the vehicle itself, and obtain the length of the candidate lane region; Obtain the third product result of the speed of each obstacle on the candidate lane region and the corresponding deviation angle; Based on the first ratio of the sum of the third product results of all obstacles and the length of the candidate lane region, determine the dynamic density index of the candidate lane region; Obtain the gain coefficient of the traffic flow cost; Determine the cost of the traffic flow of the candidate lane region according to the second ratio of the gain coefficient of the traffic flow cost and the dynamic density index of the candidate lane.

9. The driving method of the vehicle in the shunt section according to claim 5, characterized in that, The cost of obtaining the obstacles in the candidate lane region includes: Obtain the third ratio of the area of each obstacle to the area of the candidate lane region; Obtain the fourth product result of the sum of the third ratios corresponding to all obstacles and the obstacle gain coefficient; wherein, the obstacle gain coefficient has a positive correlation with the number of obstacles on the candidate lane region; Take the fourth product result as the cost of the obstacles in the candidate lane region.

10. The driving method of the vehicle on the shunt section according to claim 5, characterized in that, The cost of obtaining the navigation distance of the candidate lane region includes: Obtain the minimum distance value between the navigation distance of the candidate lane region and the distance threshold; Obtain the fifth product result of the minimum distance value and the navigation distance gain coefficient; wherein, the navigation distance gain coefficient when the navigation direction is non - straight is greater than the navigation distance gain coefficient when the navigation direction is straight; Take the negative value of the fifth product result as the cost of the navigation distance of the candidate lane region.

Images