Speed planning method and device for autonomous vehicle, electronic equipment and medium

Abstract

The invention provides a speed planning method and device for an autonomous vehicle, electronic equipment and a medium. The speed planning method comprises the following steps: clustering historical driving tracks of a target vehicle in an afflux area in front of the autonomous vehicle to generate a traffic flow guide line of the afflux area; generating a target vehicle guide line according to driving information of a target vehicle and the traffic flow guide line; clustering the historical speed of the target vehicle in the confluence area to generate the confluence traffic flow speed of the target vehicle; and processing the afflux traffic flow speed and the target vehicle guide line through a decision algorithm to obtain the target speed of the own vehicle so as to solve the problem that the speed of the own vehicle cannot be accurately planned.

Description

Technical Field

[0001] The embodiments of the present application relate to the field of autonomous driving technology, and in particular to a speed planning method, device, electronic device, and medium for an autonomous driving vehicle. Background Art

[0002] In the practical application of autonomous driving, vehicles primarily rely on navigation systems to determine their location and route. However, when a vehicle enters a complex intersection, such as one with irregular lane markings, unclear lane markings, or the absence of a map, the navigation system's predicted trajectory is inaccurate, making it impossible to plan the vehicle's speed and thus accurately guide driving behavior, thus compromising the safety and reliability of autonomous driving. Summary of the Invention

[0003] The embodiments of the present application provide a speed planning method, device, electronic device, and medium for an autonomous driving vehicle, aiming to solve the problem of being unable to accurately plan the speed of the vehicle.

[0004] To solve the above problems, the present application discloses a speed planning method for an autonomous vehicle, comprising:

[0005] Clustering the historical driving trajectories of target vehicles in the merging area ahead of the vehicle to generate a traffic flow guide line for the merging area;

[0006] generating the target vehicle guide line according to the driving information of the target vehicle and the traffic flow guide line;

[0007] Clustering the historical speeds of the target vehicles in the merging area to generate a merging traffic speed of the target vehicles;

[0008] The incoming vehicle flow speed and the target vehicle guide line are processed by a decision algorithm to obtain the target speed of the own vehicle.

[0009] According to the above technical means, the present application regenerates the traffic guide lines and traffic speeds of the merging area by clustering the historical trajectories and historical speeds of the target vehicles, and then generates the target vehicle guide lines based on the driving information of the target vehicles and the traffic guide lines, thereby avoiding the impact of unclear lane lines or lack of maps on autonomous driving. Then, the merging traffic speed and the target vehicle guide lines are processed by a decision algorithm to obtain the target speed of the vehicle. In this way, a more reasonable and effective target speed of the vehicle can be planned in a complex intersection environment, thereby improving the reliability and safety of autonomous driving.

[0010] Optionally, clustering the historical driving trajectories of target vehicles in the merging area ahead of the vehicle to generate the traffic guide line of the merging area includes: clustering the historical driving trajectories of the target vehicles in the merging area by a clustering algorithm to obtain multiple historical trajectory clusters, each of the historical trajectory clusters including: a cluster center; and fitting the cluster center of each historical trajectory cluster to obtain the traffic guide line of the merging area.

[0011] According to the above technical means, the driving trajectory of the target vehicle can be adjusted in advance through the traffic guide lines in the merging area, and the risk of collision between the vehicle and the target vehicle can also be reduced.

[0012] Optionally, the decision algorithm is used to process the merging traffic speed and the target vehicle guide line to obtain the target speed of the own vehicle, including: determining the merging point of the own vehicle and the target vehicle based on the traffic guide line and the lane line of the own vehicle; determining the relative speed between the cruising speed of the own vehicle and the merging traffic speed; determining the risk speed difference factor between the target vehicle and the own vehicle based on the current speed of the own vehicle, the merging traffic speed and the distance between the own vehicle and the merging point; determining the target speed based on the merging traffic speed, the relative speed, the risk speed difference factor and calibration parameters, wherein the calibration parameters include: a merging speed weight, a speed weight set by the own vehicle and a risk correction weight.

[0013] According to the above technical means, the target speed of the vehicle is adjusted according to the merging speed of the target vehicle, which can significantly reduce the risk of rear-end or side collision caused by speed difference and improve the safety of autonomous driving.

[0014] Optionally, the method also includes: judging the number of target vehicles in the merging area; if the number of target vehicles is greater than a set threshold, executing the step of clustering the historical driving trajectories of target vehicles in the merging area in front of the own vehicle to generate a traffic flow guide line for the merging area; if the number of target vehicles is equal to the set threshold, determining the merging point of the own vehicle and the target vehicles based on the traffic flow guide line and the lane line of the own vehicle; calculating the first time when the own vehicle arrives at the merging point; calculating the second time when the target vehicle arrives at the merging point; determining the time difference between the own vehicle and the target vehicle arriving at the merging point based on the first time and the second time; if the time difference is less than a risk time threshold, determining that there is a collision risk between the own vehicle and the target vehicle; if the time difference is greater than or equal to the risk time threshold, determining that there is no collision risk between the own vehicle and the target vehicle.

[0015] According to the above technical means, different methods are used to determine whether the ego vehicle and the target vehicle will collide according to the number of target vehicles entering the merging area, thereby ensuring the safety of the ego vehicle in different driving scenarios and improving the reliability of autonomous driving.

[0016] Optionally, the method further includes: determining a merging interval between the ego vehicle and the target vehicle; calculating a comprehensive cost value of the ego vehicle's current speed and the target speed at each planning point within the merging interval using a cost function; and adjusting the target speed of the ego vehicle according to a speed trajectory corresponding to the comprehensive cost value.

[0017] According to the above technical means, the target speed of the vehicle can be slowly adjusted, thereby improving the comfort of autonomous driving.

[0018] In order to solve the above problems, the present application also discloses a speed planning device for an autonomous driving vehicle, comprising: a first clustering module, which clusters the historical driving trajectories of target vehicles in the merging area in front of the ego vehicle to generate a traffic guide line for the merging area; a generation module, which generates the target vehicle guide line based on the driving information of the target vehicle and the traffic guide line; a second clustering module, which clusters the historical speeds of the target vehicles in the merging area to generate the merging traffic speed of the target vehicle; and a decision module, which processes the merging traffic speed and the target vehicle guide line through a decision algorithm to obtain the target speed of the ego vehicle.

[0019] Optionally, the first clustering module includes: a historical trajectory clustering unit, configured to cluster the historical driving trajectories of the target vehicles in the merging area using a clustering algorithm to obtain a plurality of historical trajectory clusters, each of the historical trajectory clusters including: a cluster center; and a fitting unit, configured to fit the cluster center of each historical trajectory cluster to obtain the traffic guide line of the merging area.

[0020] Optionally, the decision module includes: a merging point unit, used to determine the merging point of the ego vehicle and the target vehicle based on the target vehicle guide line and the lane line of the ego vehicle; a relative speed unit, used to determine the relative speed of the ego vehicle and the merging traffic speed; a calculation unit, used to determine the risk speed difference factor between the target vehicle and the ego vehicle based on the current speed of the ego vehicle, the merging traffic speed and the distance between the ego vehicle and the merging point; a determination unit, used to determine the target speed based on the merging traffic speed, the relative speed, the risk speed difference factor and calibration parameters, wherein the calibration parameters include: a merging speed weight, a speed weight set by the ego vehicle and a risk correction weight.

[0021] In order to solve the above problems, the present application also discloses an electronic device, including a processor and a memory, wherein

[0022] Memory for storing computer programs;

[0023] The processor is used to execute the program stored in the memory to implement the speed planning method of the autonomous driving vehicle.

[0024] In order to solve the above problems, the present application also discloses a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, it implements the speed planning method of an autonomous driving vehicle.

[0025] Of course, any product implementing the present application does not necessarily need to achieve all of the advantages described above at the same time. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1 This is a flow chart of a speed planning method for an autonomous driving vehicle provided in one embodiment of the present application;

[0027] Figure 2 is a schematic diagram of the historical trajectory of the target vehicle provided in an embodiment of the present application;

[0028] Figure 3 This is a flow chart of a speed planning method for an autonomous driving vehicle provided in one embodiment of the present application;

[0029] Figure 4 is a schematic diagram of a vehicle and a target vehicle merging in a merging area according to an embodiment of the present application;

[0030] Figure 5 A flow chart of a speed planning method for an autonomous driving vehicle provided in one embodiment of the present application;

[0031] Figure 6 is a target speed diagram generated using the speed planning method for the autonomous driving vehicle of the present application;

[0032] Figure 7 This is a cost diagram provided by an embodiment of the present application;

[0033] Figure 8 This is a structural diagram of a communication signal quality evaluation device provided in an embodiment of the present application;

[0034] Figure 9 This is a structural diagram of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0035] In order to make the technical problems, technical solutions and beneficial effects solved by this application more clearly understood, this application is further described in detail below in conjunction with the embodiments. It should be understood that the specific embodiments described herein are only used to explain this application and are not intended to limit this application.

[0036] Example 1

[0037] refer to Figure 1 , shows a speed planning method for an autonomous driving vehicle provided by an embodiment of the present application, comprising the following steps:

[0038] Step 101: clustering the historical driving trajectories of target vehicles in the merging area ahead of the vehicle to generate a traffic flow guide line for the merging area.

[0039] Step 102: Generate the target vehicle guide line according to the driving information of the target vehicle and the traffic flow guide line.

[0040] Step 103: Clustering the historical speeds of the target vehicles in the merging area to generate the merging speed of the target vehicles.

[0041] Step 104: Process the incoming vehicle flow speed and the target vehicle guide line through a decision algorithm to obtain the target speed of the own vehicle.

[0042] In this embodiment, the historical trajectories and historical speeds of target vehicles are clustered to regenerate the traffic flow guide lines and traffic flow speeds in the merging area. Target vehicle guide lines are then generated based on the target vehicle's driving information and traffic flow guide lines, thereby avoiding the impact of unclear lane lines or the lack of a map on autonomous driving. The merging traffic flow speeds and target vehicle guide lines are then processed by a decision algorithm to obtain the target speed of the ego vehicle. This allows for a more reasonable and effective target speed of the ego vehicle to be planned even in complex intersection environments, thereby improving the reliability and safety of autonomous driving.

[0043] In step 101, the merging area in front of the vehicle generally refers to a merging area or a merging area. Merging or merging refers to the process of different roads or lanes merging into the same traffic flow, such as: highway entrance ramps, construction areas, bridge entrances, urban expressway merging points, etc.

[0044] In step 101, it is necessary to obtain the historical positions of multiple target vehicles in the merging area in front of the vehicle in advance, project the historical positions of the target vehicles into the vehicle coordinate system to form historical trajectory points, and then obtain the historical trajectory of the target vehicle based on the historical trajectory points.

[0045] For example: The position vector of the target vehicle P in its own coordinate system {B} is Bp, the translation position vector of the coordinate system {B} relative to the vehicle's own coordinate system {A} is A p B , the rotation transformation matrix between coordinate systems is Then the translation position vector of point P in coordinate system {A} is A p can be expressed as:

[0046] A p= B p+ A p B

[0047] Rotation transformation matrix The function is to transform the coordinate system of the point from {B} to the coordinate system of {A}. The specific formula is:

[0048]

[0049] Transformation Matrix The role of is to combine translation transformation and rotation transformation together, which can be expressed as:

[0050]

[0051] In local positioning, the upstream module can also obtain the translation amounts in the x, y, and z directions relative to the starting coordinate system at each moment, as well as the quaternion that can be used for coordinate rotation transformation. The quaternion can be converted into a rotation matrix, and all parameters in the above formula are known quantities.

[0052] For a quaternion q(w,x,y,z), convert it to a rotation matrix R:

[0053]

[0054] For the coordinate transformation of the moving target vehicle between adjacent moments, i-1 T represents the transformation matrix of the vehicle relative to the world coordinate system at time i-1, i T represents the transformation matrix of the vehicle relative to the world coordinate system at time i, then the coordinate transformation at time i-1 relative to time i is:

[0055]

[0056] Through the above calculations, we can get the historical trajectory coordinates at the current moment and the historical time. Based on the historical trajectory coordinates, the historical driving trajectory is generated. The generated historical driving trajectory is as follows: Figure 2 As shown, in Figure 2In the figure, there are two target vehicles in the merging area A in front of the ego vehicle (ego), namely target vehicle 1 (obs1) and target vehicle 2 (obs2). The historical driving trajectories are generated based on the historical trajectory coordinates of the target vehicles. Then, the historical driving trajectories of multiple target vehicles are clustered to generate the traffic flow guide lines of the merging area. Figure 2 In the figure, two target vehicles are used as an example to illustrate the process of generating historical driving trajectories. However, in actual applications, there may be multiple target vehicles entering the area.

[0057] In step 102, the driving information may include: the speed and acceleration of the target vehicle, and the target vehicle guide line, that is, the lane line of the target vehicle, may be generated according to the speed of the target vehicle, the acceleration of the target vehicle, and the traffic guide line.

[0058] In step 103, the historical speeds of the target vehicles in the merging area are clustered in the following manner to generate the merging speed of the target vehicles, including:

[0059] First, initialize the parameters and set the number of clusters K = 2 (assuming the number of lanes is 2). In practical applications, the number of clusters can be determined according to the number of lanes.

[0060] Randomly select K initial cluster centers centroids[k] from the historical speed dataset (speed_data) of the target vehicle.

[0061] Secondly, iterative optimization is performed, repeating the following steps of assigning data points to the nearest cluster and updating the cluster center until the cluster center no longer changes (converges).

[0062] The steps to assign a data point to the nearest cluster include:

[0063] For each speed value s, the distance from each speed value s to all cluster centers is calculated using the Euclidean distance.

[0064] Assign each speed value s to the category to which the nearest cluster center belongs.

[0065] Updating cluster centers includes:

[0066] For each cluster k, calculate the average value of all speed data in the cluster;

[0067] Update the initial cluster center centroids[k] to the average value to obtain the merging vehicle speed.

[0068] In practical applications, step 104 includes the following sub-steps:

[0069] Sub-step 1041: Determine a merging point between the ego vehicle and the target vehicle based on the target vehicle guide line and the lane line of the ego vehicle.

[0070] In actual applications, the location where the ego vehicle and the target vehicle are most likely to collide is generally at the intersection of the target vehicle and the ego vehicle. Therefore, this embodiment determines the meeting point of the ego vehicle and the target vehicle based on the target vehicle guide line and the lane line of the ego vehicle.

[0071] Sub-step 1042: Determine the relative speed between the vehicle and the merging vehicle.

[0072] In practical applications, the cruising speed of the vehicle can be obtained in advance, and the cruising speed of the vehicle can be subtracted from the speed of the merging traffic flow to obtain the relative speed of the cruising speed of the vehicle with respect to the speed of the merging traffic flow.

[0073] Sub-step 1043: Determine a risk speed difference factor between the target vehicle and the ego vehicle based on the current speed of the ego vehicle, the speed of the merging vehicle, and the distance between the ego vehicle and the merging point.

[0074] In practical applications, the risk speed difference factor is determined based on the cruising speed of the vehicle, the speed of the merging vehicle, and the distance between the vehicle and the merging point. The specific calculation formula is as follows:

[0075] ΔVrisk=(Vego-Vmerge) / S

[0076] Wherein, ΔVrisk represents the risk speed difference factor, Vego represents the current speed of the ego vehicle, Vmerge represents the speed of the merging vehicle, and S represents the distance between the ego vehicle and the merging point.

[0077] Sub-step 1044: Determine the target speed of the ego vehicle based on the merging vehicle speed, the relative speed, the risk speed difference factor, and calibration parameters, wherein the calibration parameters include: a merging speed weight, a speed weight set by the ego vehicle, and a risk correction weight.

[0078] In practical applications, the target speed of the vehicle can be calculated using the following formula:

[0079] Vtarget=α·Vmerge+β·(Vset-Vmerge)+γ·ΔVrisk

[0080] Where Vtarget represents the target speed of the ego vehicle, Vmerge represents the merging speed, Vset represents the cruising speed of the ego vehicle, ΔVrisk represents the risk speed difference factor, α represents the merging speed weight, which mainly reflects the following priority of the ego vehicle in the traffic flow, β represents the speed weight set by the ego vehicle, which mainly reflects the driver's preference, and γ represents the risk adjustment weight, which mainly reflects the sensitivity of safety requirements.

[0081] By determining the target speed of the ego vehicle based on the incoming vehicle flow speed, the relative speed, the risk speed difference factor, and calibration parameters, the risk of collision between the ego vehicle and the target vehicle can be avoided.

[0082] Example 2

[0083] refer to Figure 3 , shows a speed planning method for an autonomous driving vehicle provided by an embodiment of the present application, comprising the following steps:

[0084] Step 301: Determine whether the number of target vehicles in the merging area is greater than a set threshold. If the number of target vehicles is greater than the set threshold, execute step 302; if the number of target vehicles is equal to the set threshold, execute step 306.

[0085] In practical applications, the current scene information can be obtained in advance. The scene information includes: merging scenes and merging scenes, and then the type of scene information is determined. When it belongs to a merging scene or a merging scene, when determining the merging area in front of the vehicle, since it is mainly moving targets that pose a risk to the driving of the vehicle, it is necessary to filter out the stationary targets in the merging area and only count the number of moving target vehicles.

[0086] The number of target vehicles in the merging area is determined. If the number of target vehicles is greater than a set threshold, the historical trajectories of the target vehicles are obtained, and the historical trajectories are clustered to generate traffic flow guidance lines for the merging area. If the number of target vehicles is equal to the set threshold, it is directly determined whether there is a collision risk between the vehicle and the target vehicles.

[0087] Among them, the threshold value can be set by a person skilled in the art in any appropriate manner, such as setting the threshold value by manual experience, or setting the threshold value based on the difference value of historical data. Preferably, in actual applications, the threshold value can be set to 1, and this application does not impose any restrictions on this.

[0088] Step 302: Clustering the historical driving trajectories of target vehicles in the merging area ahead of the vehicle to generate a traffic flow guide line for the merging area.

[0089] In practical applications, step 302 includes the following sub-steps:

[0090] Sub-step 3021: Clustering the historical driving trajectories of the target vehicles entering the merging area using a clustering algorithm to obtain a plurality of historical trajectory clusters, each of which includes a cluster center.

[0091] Sub-step 3022: Fit the cluster center of each historical trajectory cluster to obtain the traffic flow guide line of the merging area.

[0092] In practical applications, K-means clustering can be used to cluster the historical trajectories of target vehicles entering the area, and ultimately obtain the traffic flow guidance lines of the merging area. The specific clustering process is as follows:

[0093] (1) The cluster center K is determined according to the number of target vehicles in the merging area, and the historical trajectory clusters are determined according to the number of cluster centers. That is, there are as many cluster centers as there are target vehicles in the merging area, and there are as many historical trajectory clusters as there are cluster centers. For example, if the number of target vehicles is 4, there are 4 cluster centers and 4 historical trajectory clusters.

[0094] (2) Initialize K cluster centers.

[0095] (3) Repeat the following iterative process until the stopping condition is met.

[0096] The Euclidean distance between the historical position point of each target vehicle and each cluster center is calculated, and the historical position point of each target vehicle is assigned to the historical trajectory cluster described by the cluster center closest to it.

[0097] The distance d between the historical location point of each target vehicle and each cluster center is calculated using the following formula.

[0098]

[0099] X: represents the historical location point of a target vehicle. Ci represents the i-th cluster center.

[0100] d(X,Ci) represents the distance from the historical location point X to the cluster center Ci.

[0101] distance(.) represents a specific distance calculation function, which can be Euclidean distance, etc.

[0102] N represents the number of historical location points.

[0103] The historical trajectory cluster to which the historical location point X is assigned to the center point closest to it is calculated using the following formula.

[0104] cluster(X)=argmini{d(X,Ci)}

[0105] argminb means returning the cluster center that minimizes d(X,Ci).

[0106] Through the above operations, the historical position points of each target vehicle are assigned to each historical trajectory cluster. Then, for each historical trajectory cluster, the average value of all historical position points is calculated. Based on this average value, the cluster center is updated, that is, the average value is used as the new cluster center of each historical trajectory cluster.

[0107] The new cluster center is calculated using the following formula:

[0108]

[0109] C i : The new cluster center of the i-th cluster.

[0110] ∑ X∈ cluster ( i ) X represents the sum of all data points X belonging to the i historical trajectory clusters.

[0111] number of data points in cluster(i) represents the number of data points contained in the i-th cluster.

[0112] Through the above operation, a new cluster center of each historical trajectory cluster is obtained. The cluster center of each historical trajectory cluster is fitted to obtain the traffic flow guide line of the merging area. For example, the traffic flow guide line is y=ax 3 +bx 2 +cx.

[0113] Step 303: Generate the target vehicle guide line according to the driving information of the target vehicle and the traffic flow guide line.

[0114] Step 304: Clustering the historical speeds of the target vehicles in the merging area to generate the merging speed of the target vehicles.

[0115] Step 305: Process the incoming vehicle speed and the target vehicle guide line through a decision algorithm to obtain the target speed of the own vehicle.

[0116] Step 306: Determine a merging point between the ego vehicle and the target vehicle based on the target vehicle guide line and the ego vehicle's lane line.

[0117] Step 307: Calculate the first time when the ego vehicle arrives at the merging point.

[0118] Step 308: Calculate the second time when the target vehicle arrives at the merging point.

[0119] Step 309: Determine the time difference between the ego vehicle and the target vehicle arriving at the merging point based on the first time and the second time.

[0120] Step 310: Determine whether the time difference is less than the risk time threshold. If the time difference is less than the risk time threshold, execute step 311; if the time difference is greater than or equal to the risk time threshold, execute step 312.

[0121] Step 311: Determine whether there is a collision risk between the ego vehicle and the target vehicle.

[0122] Step 312: Determine that there is no collision risk between the ego vehicle and the target vehicle.

[0123] In a specific application, the first time T_reach_ego, that is, the estimated time for the ego vehicle to reach the merging point, is calculated. The second time T_reach_obs, that is, the estimated time for the target vehicle to reach the merging point, is calculated. The time difference T_diff between the ego vehicle and the target vehicle reaching the merging point is determined based on the first time and the second time, and the T_diff is compared with the risk time threshold T_risk. If T_diff is less than T_risk, it means that the merging area belongs to a high-risk area, that is, there is a risk of collision between the ego vehicle and the target vehicle, and therefore, immediate avoidance is required. If T_diff is equal to T_risk, it means that the merging area belongs to a medium-risk area, that is, there is no risk of collision between the ego vehicle and the target vehicle, and continuous monitoring is required. If T_diff is greater than T_risk, it means that the merging area belongs to a low-risk area, that is, there is no risk of collision between the ego vehicle and the target vehicle, and the ego vehicle maintains its current path.

[0124] The risk time threshold can be calculated using the following formula:

[0125]

[0126] Among them, ΔV represents the speed difference between the vehicle and the target vehicle. The larger the speed difference, the higher the risk of rear-end collision. The smaller Δs or V ego The larger it is, the more risk increases dramatically. K is the weight coefficient, which is a constant.

[0127] In order to better understand how to determine whether the vehicle and the target vehicle collide at the merging point in the merging area, the following Figure 4 Detailed instructions, in Figure 4In the example, the merging point of the ego vehicle ego and the target vehicle obs is point A. The speed difference ΔV between the ego vehicle ego and the target vehicle obs, the distance D_merge_ego between the ego vehicle ego and the merging point A, and the distance D_merge_obs between the target vehicle obs and the merging point A are determined. The distance difference Δs between the ego vehicle ego and the target vehicle obs is determined based on D_merge_ego and D_merge_obs. The time risk threshold T_risk is obtained based on the speed difference, the weight system, the distance difference, and the speed of the ego vehicle. If T_diff is greater than T_risk, there is a risk of collision between the ego vehicle and the target vehicle, and the ego vehicle changes lanes to avoid the danger.

[0128] In this embodiment, different methods are used to determine whether the ego vehicle and the target vehicle collide according to the number of target vehicles entering the merging area, thereby ensuring the safety of the ego vehicle in different driving scenarios and improving the reliability of autonomous driving.

[0129] Example 3

[0130] refer to Figure 5 , shows a speed planning method for an autonomous driving vehicle provided by an embodiment of the present application, comprising the following steps:

[0131] Step 501: Clustering the historical driving trajectories of target vehicles in the merging area ahead of the vehicle to generate a traffic flow guide line for the merging area.

[0132] Step 502: Generate the target vehicle guide line according to the driving information of the target vehicle and the traffic flow guide line.

[0133] Step 503: Clustering the historical speeds of the target vehicles in the merging area to generate the merging speed of the target vehicles.

[0134] Step 504: Process the incoming vehicle speed and the target vehicle guide line through a decision algorithm to obtain the target speed of the own vehicle.

[0135] Step 505: Determine a merging section between the vehicle and the target vehicle.

[0136] Step 506: Calculate the comprehensive cost value of the current speed of the ego vehicle and the target speed at each planning point within the merging interval using a cost function.

[0137] Step 507: Adjust the target speed of the vehicle according to the speed trajectory corresponding to the comprehensive cost value.

[0138] In a specific application, the target speed of the host vehicle has been determined through step 504. When there is no target vehicle, the host vehicle travels at this target speed. However, when there is a target vehicle, the host vehicle needs to decelerate, or determine the deceleration value. The dynamic programming (DP) rule is introduced to plan the target speed. The schematic diagram of the target speed planned using DP is as shown in Figure 6 shown in Figure 6 where the original speed curve of the host vehicle is the motion curve when there is no target vehicle around the host vehicle, and the planned speed curve is the speed curve of the host vehicle generated through DP planning when there is a target vehicle around the host vehicle.

[0139] During the DP planning process, first, the confluence area [s0, s_merge_p] between the host vehicle and the target vehicle needs to be determined, where s0 is the starting point of the host vehicle, and s_merge_p is the position of the merging point of the host vehicle.

[0140] Within [s0, s_merge_p], starting from s0, the comprehensive cost value between the current speed of the host vehicle and the target speed at each planned point of the host vehicle is calculated through the cost function cost.

[0141] The formula for the cost function is: cost(Vtarget) = (Vego - Vtarget) 2 *WeightVtarget

[0142] Vtarget represents the target speed, Vego represents the current actual speed of the host vehicle, and Weight Vtarget represents the weight coefficient.

[0143] Select the comprehensive cost value with the minimum cost value from multiple comprehensive cost values, and then adjust the target speed of the host vehicle according to the speed trajectory corresponding to the minimum comprehensive cost value.

[0144] In Figure 7 shows the relationship between the current speed (V) and the target speed (Vtarget) of the host vehicle and the corresponding cost (cost). In Figure 7 when V < Vtarget, that is, the left area in the figure, it represents the cost when the current speed is lower than the target speed. When V > Vtarget, that is, the right area in the figure, it represents the cost when the current speed is higher than the target speed. In an ideal state, when V = Vtarget, the cost is zero, which is the optimal state.

[0145] In this embodiment, by introducing the cost function, the target speed of the host vehicle can be adjusted slowly, improving the comfort of autonomous driving.

[0146] Embodiment 4

[0147] Refer to Figure 8, showing that the embodiment of the present application further provides a speed planning device 60 for an autonomous driving vehicle, the device comprising:

[0148] The first clustering module 810 clusters the historical driving trajectories of target vehicles in the merging area ahead of the vehicle to generate a traffic flow guide line for the merging area.

[0149] The generating module 820 is configured to generate the target vehicle guide line according to the driving information of the target vehicle and the traffic flow guide line.

[0150] The second clustering module 830 is configured to cluster the historical speeds of the target vehicles in the merging area to generate the merging speed of the target vehicles.

[0151] The decision module 840 is configured to process the incoming vehicle flow speed and the target vehicle guide line through a decision algorithm to obtain the target speed of the vehicle.

[0152] Optionally, the first clustering module includes:

[0153] A historical trajectory clustering unit is used to cluster the historical driving trajectories of the target vehicles entering the merging area by using a clustering algorithm to obtain a plurality of historical trajectory clusters, each of which includes: a cluster center;

[0154] A fitting unit is used to fit the cluster center of each historical trajectory cluster to obtain a traffic flow guide line of the merging area.

[0155] Optionally, the decision module includes:

[0156] a merging point unit, configured to determine a merging point between the own vehicle and the target vehicle based on the guide line of the target vehicle and the lane line of the own vehicle;

[0157] a relative speed unit, configured to determine a relative speed between the cruising speed of the vehicle and the speed of the merging vehicle;

[0158] a calculation unit, configured to determine a risk speed difference factor between the target vehicle and the ego vehicle based on the current speed of the ego vehicle, the speed of the merging vehicle, and the distance between the ego vehicle and the merging point;

[0159] The determination unit is configured to determine the target speed based on the merging vehicle speed, the relative speed, the risk speed difference factor, and calibration parameters, wherein the calibration parameters include: a merging speed weight, a speed weight set by the own vehicle, and a risk correction weight.

[0160] Optionally, the device further comprises:

[0161] a determination module, configured to determine the number of the target vehicles entering the merging area;

[0162] If the result of the judgment module is that the number of the target vehicles is greater than the set threshold, the operation of clustering the historical driving trajectories of the target vehicles in the merging area ahead of the vehicle to generate a traffic flow guide line for the merging area is performed;

[0163] If the result of the judgment module is that the number of the target vehicles is equal to the set threshold, then determining the merging point of the own vehicle and the target vehicles according to the guide line of the target vehicles and the lane line of the own vehicle;

[0164] A first calculation module is used to calculate a first time when the ego vehicle arrives at the merging point;

[0165] A second calculation module is used to calculate a second time when the target vehicle arrives at the merging point;

[0166] A time difference module, configured to determine a time difference between the vehicle and the target vehicle arriving at the merging point based on the first time and the second time;

[0167] An output module is configured to determine that there is a collision risk between the ego vehicle and the target vehicle if the time difference is less than a risk time threshold; and to determine that there is no collision risk between the ego vehicle and the target vehicle if the time difference is greater than or equal to the risk time threshold.

[0168] The device further includes: a determination module for determining a merging section between the ego vehicle and the target vehicle;

[0169] A calculation module is configured to calculate a comprehensive cost value of the current speed of the ego vehicle and the target speed at each planning point within the merging interval using a cost function; and adjust the target speed of the ego vehicle according to a speed trajectory corresponding to the comprehensive cost value.

[0170] According to the above technical means, the target speed of the vehicle can be slowly adjusted, thereby improving the comfort of autonomous driving.

[0171] The present application also provides an electronic device 90, please refer to Figure 9 , including a processor 910 and a memory 920, wherein the memory 910 is used to store computer programs; the processor 920 is used to execute the programs stored in the memory 910 to implement a speed planning method for an autonomous driving vehicle introduced in any embodiment of the present application.

[0172] An embodiment of the present application also provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, it implements a speed planning method for an autonomous driving vehicle introduced in any embodiment of the present application.

[0173] In this application, a plurality refers to two or more.

[0174] The terms "first," "second," "third," "fourth," etc. in this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence.

[0175] The term "and / or" in this application simply describes an association between related objects, indicating that three possible relationships exist. For example, A and / or B can represent: A exists alone, A and B exist simultaneously, and B exists alone. In addition, the character " / " in this application generally indicates that the related objects are in an "or" relationship.

[0176] Unless otherwise specified, all steps of the present application may be performed sequentially or randomly. For example, a statement that the method includes steps A and B indicates that the method may include steps A and B performed sequentially, or steps B and A performed sequentially. For example, a statement that the method may also include step C indicates that step C may be added to the method in any order, for example, the method may include steps A, B, and C, or steps A, C, and B, or steps C, A, and B, etc.

[0177] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent replacements and improvements made within the spirit and principles of the present application should be included in the scope of protection of the present application.

Claims

1. A speed planning method for an autonomous driving vehicle, characterized in that: include: Clustering the historical driving trajectories of target vehicles in the merging area ahead of the vehicle to generate a traffic flow guide line for the merging area; generating the target vehicle guide line according to the driving information of the target vehicle and the traffic flow guide line; Clustering the historical speeds of the target vehicles in the merging area to generate a merging traffic speed of the target vehicles; The incoming vehicle flow speed and the target vehicle guide line are processed by a decision algorithm to obtain the target speed of the own vehicle.

2. The method according to claim 1, characterized in that The step of clustering the historical driving trajectories of target vehicles in the merging area ahead of the vehicle to generate the traffic flow guide line of the merging area includes: Clustering the historical driving trajectories of the target vehicles entering the merging area by a clustering algorithm to obtain a plurality of historical trajectory clusters, each of the historical trajectory clusters including: a cluster center; The cluster center of each historical trajectory cluster is fitted to obtain a traffic flow guide line of the merging area.

3. The method according to claim 1, characterized in that The process of processing the incoming vehicle flow speed and the target vehicle guide line by a decision algorithm to obtain the target speed of the vehicle includes: Determining a merging point between the own vehicle and the target vehicle based on the target vehicle guide line and the lane line of the own vehicle; Determining the relative speed between the cruising speed of the vehicle and the speed of the merging vehicle; Determining a risk speed difference factor between the target vehicle and the ego vehicle based on the current speed of the ego vehicle, the speed of the merging vehicle, and the distance between the ego vehicle and the merging point; The target speed is determined based on the merging vehicle speed, the relative speed, the risk speed difference factor, and calibration parameters, wherein the calibration parameters include: a merging speed weight, a speed weight set by the own vehicle, and a risk correction weight.

4. The method according to claim 1, wherein The method further comprises: Determining the number of the target vehicles entering the merging area; If the number of the target vehicles is greater than a set threshold, performing the step of clustering the historical driving trajectories of the target vehicles in the merging area ahead of the vehicle to generate a traffic flow guide line for the merging area; If the number of the target vehicles is equal to a set threshold, determining a merging point between the ego vehicle and the target vehicles based on the guide lines of the target vehicles and the lane line of the ego vehicle; Calculating the first time when the ego vehicle arrives at the merging point; Calculating a second time when the target vehicle arrives at the merging point; Determine a time difference between the vehicle and the target vehicle arriving at the merging point based on the first time and the second time; If the time difference is less than the risk time threshold, it is determined that there is a collision risk between the ego vehicle and the target vehicle; If the time difference is greater than or equal to the risk time threshold, it is determined that there is no collision risk between the ego vehicle and the target vehicle.

5. The method according to claim 1, wherein The method further comprises: determining a merging section between the vehicle and the target vehicle; Calculating a comprehensive cost value of the current speed of the vehicle and the target speed at each planning point within the merging interval using a cost function; The target speed of the vehicle is adjusted according to the speed trajectory corresponding to the comprehensive cost value.

6. A speed planning device for an autonomous driving vehicle, characterized in that: include: A first clustering module clusters the historical driving trajectories of target vehicles in the merging area ahead of the vehicle to generate a traffic flow guide line for the merging area; A generating module, configured to generate the target vehicle guide line according to the driving information of the target vehicle and the traffic flow guide line; a second clustering module, configured to cluster the historical speeds of the target vehicles in the merging area to generate a merging speed of the target vehicles; The decision module is used to process the incoming vehicle flow speed and the target vehicle guide line through a decision algorithm to obtain the target speed of the own vehicle.

7. The device according to claim 6, characterized in that The first clustering module includes: A historical trajectory clustering unit is used to cluster the historical driving trajectories of the target vehicles entering the merging area by using a clustering algorithm to obtain a plurality of historical trajectory clusters, each of which includes: a cluster center; A fitting unit is used to fit the cluster center of each historical trajectory cluster to obtain the traffic flow guide line of the merging area.

8. The device according to claim 6, characterized in that The decision module includes: a merging point unit, configured to determine a merging point between the own vehicle and the target vehicle based on the guide line of the target vehicle and the lane line of the own vehicle; a relative speed unit, configured to determine a relative speed between the cruising speed of the vehicle and the speed of the merging vehicle; a calculation unit, configured to determine a risk speed difference factor between the target vehicle and the ego vehicle based on the current speed of the ego vehicle, the speed of the merging vehicle, and the distance between the ego vehicle and the merging point; The determination unit is configured to determine the target speed based on the merging vehicle speed, the relative speed, the risk speed difference factor, and calibration parameters, wherein the calibration parameters include: a merging speed weight, a speed weight set by the own vehicle, and a risk correction weight.

9. An electronic device, characterized in that: comprising a processor and a memory, wherein Memory for storing computer programs; A processor is used to execute a program stored in a memory to implement a speed planning method for an autonomous driving vehicle as described in any one of claims 1 to 5.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, it implements the speed planning method for an autonomous driving vehicle according to any one of claims 1 to 5.

Images