Speed decision method, system, vehicle and device for vehicle

Abstract

The application discloses a speed decision method and system of a vehicle, a vehicle and equipment. The speed decision method of the vehicle comprises the following steps: obtaining a dynamically planned speed curve; performing longitudinal decision on each obstacle according to the speed curve, wherein the longitudinal decision comprises a following strategy, an overtaking strategy and an ignoring strategy; searching for a target obstacle, wherein the target obstacle comprises an obstacle selected from the obstacles adopting the following strategy; and performing in-lane forward longitudinal simulation on the ego vehicle according to the target obstacle to obtain a final speed curve. By using the application, the speed curve (coarse solution) obtained by dynamic planning can be simulated in a lane forward manner, so that the pressure on a downstream speed planning module can be reduced, the solution speed of speed planning is improved, in addition, in the lane forward simulation, the end point constraint can be followed, and the solution speed of speed planning can be effectively improved and the solution of scenes such as emergency braking and cutting in can be ensured.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of vehicles, in particular to a speed decision method and system of a vehicle, a vehicle and equipment. BACKGROUND

[0002] Computer technology, information technology, automatic control technology and artificial intelligence technology are gradually applied in the field of automatic driving technology. Vehicle speed decision technology is one of the key technologies of automatic driving, and is also a basic problem and necessary condition for researching intelligent vehicles. Vehicle speed decision technology is to calculate a speed curve that meets safety, comfort and feasibility according to the state of the ego vehicle and environmental information. At present, the commonly used speed decision method is a method based on dynamic programming, which samples on the ST graph, calculates all feasible solutions of the entire time in the forward direction, searches for the result with the minimum cost in the reverse direction, and finally outputs the speed curve, that is, the rough solution of speed planning.

[0003] The following technical defects exist: (1) The speed curve obtained based on dynamic programming is relatively rough, the speed curve is not smooth and has a broken line, and the smoothness of the speed curve depends entirely on the downstream module of speed planning (speed planning), the objective function of speed planning needs to be solved between smoothness and constraints, and also needs to be consistent with the rough solution of speed as much as possible, which is too high a requirement for speed planning and is difficult to achieve in practice; (2) It is difficult to impose constraints on the end state based on dynamic programming for speed decision, which may easily lead to solution failure in some sudden braking, cutting-in and other scenarios. SUMMARY

[0004] Therefore, it is necessary to provide a speed decision method and system of a vehicle, a vehicle and equipment in view of the above technical problems, which can perform lane forward simulation on the speed curve (rough solution) obtained by dynamic programming, so as to reduce the pressure on the downstream speed planning module, improve the solution speed of speed planning, and in the lane forward simulation, follow the end constraints, effectively improve the speed planning solution speed and ensure the solution of sudden braking, cutting-in and other scenarios.

[0005] In a first aspect, a speed decision method of a vehicle is provided, comprising:

[0006] obtaining a speed curve of dynamic programming of a vehicle;

[0007] performing longitudinal decision on each obstacle according to the speed curve, wherein the longitudinal decision includes a following strategy, an overtaking strategy and an ignoring strategy;

[0008] finding a target obstacle, wherein the target obstacle includes an obstacle selected from the obstacles making the following strategy;

[0009] According to the target obstacle, a longitudinal simulation of the ego vehicle in a forward lane is performed to obtain a final speed curve.

[0010] In some examples, the longitudinal decision making for each obstacle according to the speed curve comprises:

[0011] A minimum value and a maximum value of the s direction of the minimum time point of the obstacle on the ST graph are obtained;

[0012] The minimum value and the maximum value of the s direction of the minimum time point of the obstacle on the ST graph are compared with the position of the ego vehicle in the s direction at the minimum time point, wherein:

[0013] When the position of the ego vehicle in the s direction at the minimum time point is greater than the maximum value, an overtaking decision is made for the obstacle;

[0014] When the position of the ego vehicle in the s direction at the minimum time point is less than the minimum value, or the sum of the position of the ego vehicle in the s direction at the minimum time point and the length of the ego vehicle is less than the maximum value, a following decision is made for the obstacle, otherwise an ignore strategy is made for the obstacle.

[0015] In some examples, the target obstacle is found by:

[0016] Each obstacle making a following decision is traversed to find an obstacle closest to the ego vehicle as the target obstacle.

[0017] In some examples, the method further comprises:

[0018] If there is no obstacle making a following decision, a virtual obstacle is constructed, and the virtual obstacle is taken as the target obstacle, wherein the distance between the position of the virtual obstacle and the ego vehicle is greater than a preset distance and the speed of the virtual obstacle is a reference speed of the ego vehicle.

[0019] In some examples, the longitudinal simulation of the ego vehicle in a forward lane according to the target obstacle to obtain a final speed curve comprises:

[0020] At each time, an expected distance between the ego vehicle and the target obstacle is calculated;

[0021] Based on the expected distance, an acceleration of an improved intelligent driver model is calculated, and an acceleration of an equal acceleration heuristic model is calculated;

[0022] The acceleration of the improved intelligent driver model and the acceleration of the equal acceleration heuristic model are compared to obtain an acceleration at time t;

[0023] According to the acceleration at time t, the ego vehicle position and the target obstacle position at next time t+Δt are updated, so as to obtain the final speed curve according to the updating result.

[0024] In some examples, the obtaining the speed curve of the vehicle dynamic planning comprises:

[0025] The speed curve of the vehicle dynamic planning is obtained by sampling on the ST graph, calculating a plurality of feasible solutions in a forward direction, searching for a minimum cost value of each feasible solution in a reverse direction, and taking a speed curve corresponding to the minimum cost value as the speed curve of the vehicle dynamic planning.

[0026] In a second aspect, a speed decision system of a vehicle is provided, comprising:

[0027] An obtaining module is configured to obtain a speed curve of a vehicle dynamic planning;

[0028] A longitudinal decision module is configured to make a longitudinal decision for each obstacle according to the speed curve, wherein the longitudinal decision comprises a following strategy, a passing strategy, and an ignoring strategy.

[0029] A searching module is configured to search for a target obstacle, wherein the target obstacle comprises an obstacle selected from obstacles making the following strategy.

[0030] A speed decision module is configured to make an in-lane forward longitudinal simulation for the ego vehicle according to the target obstacle, to obtain a final speed curve.

[0031] In a third aspect, a vehicle is provided, comprising the speed decision system of the vehicle according to the second aspect.

[0032] In a fourth aspect, a computer device is provided, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the speed decision method of the vehicle according to the first aspect and any possible implementation manner of the first aspect.

[0033] In a fifth aspect, a computer readable storage medium is provided, which stores a computer program, and the program is executable on a processor to implement the steps of the speed decision method of the vehicle according to the first aspect and any possible implementation manner of the first aspect.

[0034] In a sixth aspect, a computer program product is provided, which stores a computer program, and the program is executable on a processor to implement the steps of the speed decision method of the vehicle according to the first aspect and any possible implementation manner of the first aspect.

[0035] According to the embodiments of the present application, longitudinal decision can be made for each obstacle according to the speed curve of dynamic programming, after the decision, the target obstacle is searched according to the decision result, and finally the self-vehicle is simulated forward in the lane according to the target obstacle, so as to obtain the final speed curve. The speed curve (coarse solution) obtained by dynamic programming can be simulated forward in the lane, so as to reduce the pressure of the downstream speed planning module, improve the solving speed of the speed planning, and in the lane forward simulation, the end point constraint can be followed, so as to effectively improve the solving speed of the speed planning and ensure the solution of the scene such as emergency braking and cutting in. BRIEF DESCRIPTION OF DRAWINGS

[0036] Other features, objects and advantages of the present application will become more apparent from the following detailed description of non-limiting embodiments thereof, read in conjunction with the accompanying drawings:

[0037] Figure 1 A flowchart of a speed decision method of a vehicle provided by the embodiments of the present application;

[0038] Figure 2 A state diagram after initialization of an ST graph;

[0039] Figure 3 A diagram of a downward movement feasible region of each sampling point in the ST graph;

[0040] Figure 4 A diagram of reverse search;

[0041] Figure 5 A structural block diagram of a speed decision system of a vehicle provided by the embodiments of the present application;

[0042] Figure 6 A structural block diagram of a computer device provided by the embodiments of the present application. DETAILED DESCRIPTION

[0043] The present application will be further described below in conjunction with embodiments and drawings. It can be understood that the specific embodiments described herein are only used to explain the related application, but not to limit the application. In addition, it should be noted that, for the convenience of description, only the parts related to the application are shown in the drawings.

[0044] It should be noted that the features of the embodiments in the present application, i.e. the features of the embodiments, can be combined with each other without conflict. The present application will be described in detail below with reference to the drawings and in conjunction with the embodiments.

[0045] The speed decision method, system, device and medium of a vehicle according to the embodiments of the present application are described in detail below in conjunction with the drawings.

[0046] Figure 1is a flow chart of a speed decision method of a vehicle according to an embodiment of the present application. As shown in Figure 1 The speed decision method of the vehicle according to the embodiment of the present application comprises the following steps:

[0047] S101: Obtain a speed curve of dynamic planning of the vehicle.

[0048] In an embodiment of the present application, the speed curve of the vehicle can be obtained by a dynamic planning method, for example: obtaining the speed curve of dynamic planning of the vehicle comprises: calculating a plurality of feasible solutions for the entire time by sampling on the ST graph in a forward direction, searching for a minimum generation value of each feasible solution in the plurality of feasible solutions in a reverse direction, and taking the speed curve corresponding to the minimum generation value as the speed curve of dynamic planning of the vehicle.

[0049] Specifically, first, the ST graph is sampled, and each point is represented as [t, s].

[0050] In the S direction, dense sampling is used in the near distance and sparse sampling is used in the far distance. The division of the near distance and the far distance is realized by setting an S threshold, which is usually between 10m and 40m. The sampling length of S s end The path length is usually taken.

[0051] In the T direction, equidistant sampling is used, and the sampling interval is a unit time (usually 1s). The sampling length of T t end The planning time is usually taken. Wherein, the state of the ST graph after initialization is as shown in Figure 2

[0052] The cost of each sampling point [t, s] in the ST graph is set to infinity, the speed of the initial point t=0, s=0 is set to the initial speed, and the speed of each sampling point [t, s] in the ST graph is limited.

[0053] Forward recursion

[0054] The initial point t=0, s=0 is started for forward recursion, the cost of the initial point is set to 0, and each feasible sampling point [t, s] is traversed.

[0055] As shown in Figure 3 The next feasible region of each point is calculated as follows:

[0056] Δs min = s + vΔt + 0.5a min Δt 2

[0057] Δs min = s + vΔt + 0.5a max Δt 2 ​

[0058] where Δt is the time error of the current point and the next point, Δs min and Δs max are the minimum and maximum values that can be reached by the sampling point [t, s] in Δt, v is the speed of the current point, a min and a max are the minimum and maximum acceleration of the vehicle at the current point.

[0059] The state transition equation is as follows:

[0060] C(t+Δt, s+Δs) = C(t, s) + C obs (t+Δt, s+Δs) + C edge (t, s, t+Δt, s+Δs)

[0061] where C(t, s) is the cost of the point [t, s], C(t+Δt, s+Δs) is the cost of the point [t+Δt, s+Δs], C obs (t+Δt, s+Δs) is the obstacle cost of the point [t+Δt, s+Δs], and C edge (t, s, t+Δt, s+Δs) is the movement cost from the point [t, s] to the point [t+Δt, s+Δs].

[0062] The calculation expression of the obstacle cost is as follows:

[0063]

[0064] s follow = v·T follow + s follow-safe

[0065] s overtake = v obs ·T overtake + s overtake-safe

[0066] where C obs (t, s) is the obstacle cost of the point [t, s], s obs-lower and s obs-upper are the minimum and maximum values of the obstacle in the s direction, ω obs is the weight coefficient of the obstacle cost, s follow is the safety distance for following the vehicle, s overtake is the safety distance for overtaking the vehicle, v is the speed of the ego vehicle, v obs is the speed of the obstacle, T follow is the time moment for following the vehicle, T overtake is the time moment for overtaking the vehicle, s follow-safeThe minimum safety distance for following a vehicle, s overtake-safe The minimum safety distance for overtaking a vehicle.

[0067] From the point [t, s] to the point [t + Δt, s + Δs], the velocity, acceleration, jerk of the point [t + Δt, s + Δs] can be obtained, and the calculation expression of the moving cost is as follows:

[0068] C edge (t, s, t + Δt, s + Δs) = C speed + C acc + C jerk

[0069] C acc = ω acc a 2

[0070] C jerk = ω jerk j 2

[0071] Wherein, C speed is the velocity cost, C acc is the acceleration cost, and C jerk is the jerk cost.

[0072] The calculation expression of the velocity cost is as follows:

[0073]

[0074] Wherein, Δv is the velocity change from the point [t, s] to the point [t + Δt, s + Δs], v ref is the reference velocity, ω v-up and ω v-lower are the velocity change weight coefficients in different cases, ω re f is the error weight coefficient of the reference velocity, v limit is the speed limit.

[0075] The calculation expression of the acceleration cost is as follows:

[0076]

[0077] Wherein, a is the acceleration from the point [t, s] to the point [t + Δt, s + Δs], a min and a max are the minimum and maximum values of the acceleration, ω acc is the weight coefficient when the acceleration is positive, and ω dec is the weight coefficient when the acceleration is negative.

[0078] The calculation expression of the jerk cost is as follows:

[0079]

[0080] Wherein, j is the jerk from the point [t, s] to the point [t + Δt, s + Δs], j min and j max are the minimum and maximum values of the jerk, ω j-positive is the weight coefficient when the jerk is positive, and ω j-negative is the weight coefficient when the jerk is negative.

[0081] Thus, the total cost of the feasible velocity curve can be obtained by forward recursion from the initial point to the end point [t end , s end ].

[0082] Reverse search

[0083] As shown in FIG. 3, the reverse search starts from the last position sampled at s, i.e., s = s end , and the flow is as follows: Figure 4 All the sampling points at s = s end are traversed, and the position [t best , s end ] with the minimum total cost is selected, wherein t best is the time t corresponding to the sampling point with the minimum total cost;

[0084] The position [t best , s end ] with the minimum total cost starts the reverse search to find the previous point [t best - Δt, s end - Δs] with the minimum cost.

[0085] The above step is repeated until the initial point t = 0, s = 0, and the process is ended.

[0086] Since it is a reverse search, after the t-s curve is obtained, the order needs to be reversed to obtain the t-s curve in the positive order.

[0087] Conversion result

[0088] Since the velocity curve needs to know the t, s, v, and a of each point, the t-s curve obtained by the reverse search needs to be converted. Given the s and v of the initial point, the t, s, v, and a of the next point can be calculated in the forward direction by the following formula:

[0089]

[0090]

[0091] ​S102: Based on the speed curve, make longitudinal decisions for each obstacle, wherein the longitudinal decisions include a following strategy, an overtaking strategy, and an ignoring strategy.

[0092] In one embodiment of this application, longitudinal decision-making is performed on each obstacle based on the speed curve, including: obtaining the minimum and maximum values ​​of the obstacle in the s-direction at the minimum time point on the ST graph; comparing the minimum and maximum values ​​of the obstacle in the s-direction at the minimum time point on the ST graph with the position of the vehicle in the s-direction at the minimum time point, wherein: when the position of the vehicle in the s-direction at the minimum time point is greater than the maximum value, an overtaking decision is made for the obstacle; when the position of the vehicle in the s-direction at the minimum time point is less than the minimum value, or when the sum of the position of the vehicle in the s-direction at the minimum time point and the vehicle length is less than the maximum value, a following decision is made for the obstacle; otherwise, an ignore strategy is adopted for the obstacle.

[0093] Specifically, after obtaining the vehicle's speed curve, a longitudinal decision needs to be made for each obstacle (follow, overtake, ignore).

[0094] Take the minimum time t of the ST map for each obstacle. obs-min You can get the obstacle at t obs-min Minimum value s in the s direction at time s obs-min and maximum value s ego >s obs-max , and the car in t obs-min Position s in the s direction at time s ego For comparison, the length of the vehicle is l ego ,as follows:

[0095] When s ego >s obs-max When the obstacle is in the way, a longitudinal decision is made to overtake it, that is, to perform a driving operation to overtake the obstacle.

[0096] When s ego obs-min or s+l ego obs-max When the obstacle is encountered, a longitudinal decision is made to follow it, that is, to perform a driving operation to follow the obstacle.

[0097] In other cases, a longitudinal decision is made to ignore the obstacle, that is, the driving operation is not considered in relation to the obstacle.

[0098] After making longitudinal decisions about all obstacles, the next step can be taken.

[0099] S103: Locate target obstacles, wherein the target obstacles include obstacles selected from those that have adopted a follow strategy.​​

[0100] In an embodiment of the present application, the searching for the target obstacle comprises: traversing each obstacle making a following decision to find the obstacle closest to the ego vehicle as the target obstacle.

[0101] In the example, further comprising: if there is no obstacle making a following decision, constructing a virtual obstacle and taking the virtual obstacle as the target obstacle, wherein the virtual obstacle is located at a distance greater than a preset distance from the ego vehicle and has a speed of a reference speed of the ego vehicle.

[0102] For example: traversing all obstacles making a following decision to find the obstacle closest to the ego vehicle, i.e., the target obstacle. If there is no obstacle making a following decision, a virtual obstacle is constructed, which is located at a very long distance (usually set to 300 m) from the ego vehicle and has a speed of a reference speed v ref , of the ego vehicle.

[0103] S104: performing longitudinal simulation of the ego vehicle in the lane and forward according to the target obstacle to obtain a final speed curve.

[0104] In an embodiment of the present application, performing longitudinal simulation of the ego vehicle in the lane and forward according to the target obstacle to obtain a final speed curve comprises: at each time, calculating an expected distance between the ego vehicle and the target obstacle; based on the expected distance, calculating an acceleration of an improved intelligent driver model and an acceleration of an equal acceleration heuristic model; comparing the acceleration of the improved intelligent driver model and the acceleration of the equal acceleration heuristic model to obtain an acceleration at time t; updating a position of the ego vehicle and a position of the target obstacle at next time t+Δt according to the acceleration at time t, to obtain the final speed curve according to the update result.

[0105] Specifically, starting from t=0, at intervals of Δt, until t=t end , the ego vehicle is simulated longitudinally in the lane and forward to finally obtain a speed curve.

[0106] At each time, the expected distance s ref between the ego vehicle and the target obstacle is calculated as follows:

[0107]

[0108] wherein s follow-safe is a minimum safe distance for the following vehicle to travel, v is a speed of the ego vehicle, T follow is a time for the ego vehicle to travel to the target obstacle, v obs is a speed of the target obstacle, a is a comfortable acceleration of the ego vehicle, and b is a comfortable deceleration (positive number) of the ego vehicle.

[0109] Compute the acceleration a of the improved intelligent driver model (IIDM) IIDM As follows:

[0110]

[0111] where v is the ego vehicle reference speed, δ is the speed coefficient (usually taken as 4), and s is the distance between the ego vehicle and the target obstacle. ref

[0112] Compute the acceleration a of the constant acceleration heuristic model (CAH) CAH As follows:

[0113]

[0114] where ε is the step function,

[0115] Compute the acceleration a of the improved IDM model IIDM Compare with the acceleration a of the constant acceleration heuristic model CAH At time t, the acceleration is:

[0116]

[0117] where k is the acceleration proportionality coefficient. acc

[0118] According to the acceleration at time t, update the ego vehicle position and target obstacle position at the next time t+Δt, as follows:

[0119] v(t+Δt) = v(t) + aΔt

[0120] s(t+Δt) = s(t) + vΔt + 0.5aΔt 2

[0121] v obs (t+Δt) = v obs (t) + a obs Δt

[0122] s obs (t+Δt) = s obs (t) + v obs Δt + 0.5a obs Δt 2

[0123] ​​From the above formula, t, s, v, a at each moment can be obtained, so that the final result t-s curve can be composed.

[0124] According to the speed decision method of the vehicle, each obstacle can be longitudinally decided according to the speed curve obtained by dynamic programming, a target obstacle is searched according to the decision result after the decision, and finally, the vehicle is longitudinally simulated in a forward direction in a lane according to the target obstacle, so that a final speed curve is obtained. The speed curve (coarse solution) obtained by dynamic programming can be simulated in a forward direction in a lane, so that the pressure of a downstream speed planning module can be reduced, the solving speed of speed planning is improved, and in addition, in the simulation in the forward direction in the lane, the end point constraint can be followed, the solving speed of speed planning can be effectively improved, and the solution of scenes such as emergency braking and cutting in can be ensured.

[0125] Figure 5 Fig. 1 is a structural block diagram of a speed decision system of a vehicle according to an embodiment of the present application. As shown in Fig. 1, the speed decision system of the vehicle according to an embodiment of the present application comprises an acquisition module 510, a longitudinal decision module 520, a searching module 530, and a speed decision module 540, wherein: Figure 5

[0126] The acquisition module 510 is configured to obtain a speed curve of dynamic programming of the vehicle.

[0127] The longitudinal decision module 520 is configured to longitudinally decide each obstacle according to the speed curve, wherein the longitudinal decision comprises a following strategy, an overtaking strategy, and an ignoring strategy.

[0128] The searching module 530 is configured to search for a target obstacle, wherein the target obstacle comprises an obstacle selected from obstacles that make the following strategy.

[0129] The speed decision module 540 is configured to longitudinally simulate the vehicle in a forward direction in a lane according to the target obstacle, so as to obtain a final speed curve.

[0130] According to the speed decision system of the vehicle, each obstacle can be longitudinally decided according to the speed curve obtained by dynamic programming, a target obstacle is searched according to the decision result after the decision, and finally, the vehicle is longitudinally simulated in a forward direction in a lane according to the target obstacle, so that a final speed curve is obtained. The speed curve (coarse solution) obtained by dynamic programming can be simulated in a forward direction in a lane, so that the pressure of a downstream speed planning module can be reduced, the solving speed of speed planning is improved, and in addition, in the simulation in the forward direction in the lane, the end point constraint can be followed, the solving speed of speed planning can be effectively improved, and the solution of scenes such as emergency braking and cutting in can be ensured.

[0131] ​The specific definition of the vehicle speed decision system can refer to the definition of the vehicle speed decision method in the foregoing, and details are not described herein again. Each module of the vehicle speed decision system described above can be implemented by software, hardware, or a combination thereof, in whole or in part. Each module described above can be embedded in or independent of a processor in a computer device in a hardware form, or can be stored in a memory of the computer device in a software form, so that the processor invokes and executes the operation corresponding to each module.

[0132] In one embodiment, a vehicle is provided, comprising: the vehicle speed decision system according to any one of the above embodiments. The vehicle can make longitudinal decisions for each obstacle according to the dynamic programming speed curve, find a target obstacle according to the decision result after the decision, and finally make a lane-in front longitudinal simulation for the ego vehicle according to the target obstacle to obtain a final speed curve. The lane front simulation can be performed on the speed curve (coarse solution) obtained by dynamic programming, so as to reduce the pressure of the downstream speed planning module, improve the solution speed of the speed planning, and in addition, the end point constraint can be followed in the lane front simulation, so as to effectively improve the solution speed of the speed planning and ensure the solution of scenarios such as emergency braking and cut-in.

[0133] In addition, other configurations and functions of the vehicle according to the embodiments of the present application are known to those skilled in the art, and details are not described herein.

[0134] In one embodiment, a computer device is provided. Figure 6 The structural block diagram of the computer device provided in the embodiments of the present application is shown in Figure 6 . The computer device includes a memory and a processor, and the memory stores a computer program. When the processor executes the computer program, the vehicle speed decision method embodiments described above are implemented. For example, the following operations are performed: obtaining a speed curve of dynamic programming of a vehicle;

[0135] According to the speed curve, making longitudinal decisions for each obstacle, wherein the longitudinal decisions include a following strategy, an overtaking strategy, and an ignoring strategy;

[0136] Finding a target obstacle, wherein the target obstacle includes an obstacle selected from the obstacles making the following strategy;

[0137] According to the target obstacle, making a lane-in front longitudinal simulation for the ego vehicle to obtain a final speed curve.

[0138] The embodiments of the present application also provide a computer readable storage medium, which stores a computer program. When a processor executes the computer program, the vehicle speed decision method embodiments described above are implemented. For example, the following operations are performed: obtaining a speed curve of dynamic programming of a vehicle;

[0139] making longitudinal decisions for each obstacle according to the speed curve, wherein the longitudinal decisions include a following strategy, an overtaking strategy, and an ignoring strategy;

[0140] finding a target obstacle, wherein the target obstacle includes an obstacle selected from the obstacles making the following strategy;

[0141] performing in-lane forward longitudinal simulation for the ego vehicle according to the target obstacle, to obtain a final speed curve.

[0142] The embodiments of the present application provide a computer program product, which includes instructions, when the instructions are executed, causing the method as described in the embodiments of the present application to be performed. For example, each step of the speed decision method of the vehicle shown can be executed, for example, to obtain a speed curve of vehicle dynamic planning. Figure 1

[0143] making longitudinal decisions for each obstacle according to the speed curve, wherein the longitudinal decisions include a following strategy, an overtaking strategy, and an ignoring strategy;

[0144] finding a target obstacle, wherein the target obstacle includes an obstacle selected from the obstacles making the following strategy;

[0145] performing in-lane forward longitudinal simulation for the ego vehicle according to the target obstacle, to obtain a final speed curve.

[0146] A person of ordinary skill in the art can understand that all or part of the processes in the method of the above embodiments can be completed by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer readable storage medium, and when executed, can include the processes of the above embodiments. Any reference to a memory, storage, database or other medium in the embodiments provided by the present application can include at least one of a non-volatile and volatile memory. The non-volatile memory can include a read-only memory (ROM), a magnetic tape, a floppy disk, a flash memory or an optical memory. The volatile memory can include a random access memory (RAM) or an external cache memory. As an illustration but not limitation, the RAM can be in various forms, such as a static random access memory (SRAM) or a dynamic random access memory (DRAM).

[0147] ​Any technical features in the above embodiments can be combined, and for the sake of brevity, not all possible combinations of the technical features in the above embodiments are described, however, as long as the combinations of the technical features do not contradict each other, they shall be considered within the scope of the present disclosure.

[0148] The above embodiments only express several implementation manners of the present application, and the description is relatively specific and detailed, but it shall not be understood as a limitation on the patent scope of the present application. It should be noted that, for those skilled in the art, several modifications and improvements can be made without departing from the concept of the present application, and these shall be within the protection scope of the present application. Therefore, the protection scope of the patent of the present application shall be subject to the appended claims.

Claims

1. A method for determining the speed of a vehicle, characterized in that, include: Obtain the speed curve for vehicle dynamics; Based on the speed curve, longitudinal decisions are made for each obstacle, wherein the longitudinal decisions include a following strategy, an overtaking strategy, and an ignoring strategy; Find target obstacles, wherein the target obstacles include obstacles selected from those that make a follow strategy; Based on the target obstacle, a longitudinal simulation of the vehicle's forward direction within the lane is performed to obtain the final speed curve; The step of making longitudinal decisions for each obstacle based on the velocity curve includes: Obtain the minimum and maximum values ​​of the obstacle in the s direction at the minimum time point on the ST graph; The minimum and maximum values ​​of the obstacle in the s-direction at the minimum time point on the ST map are compared with the position of the vehicle in the s-direction at the minimum time point, wherein: When the position of the vehicle in the s direction at the minimum time point is greater than the maximum value, it makes an overtaking decision on the obstacle; When the position of the vehicle in the s direction at the minimum time point is less than the minimum value, or when the sum of the position of the vehicle in the s direction at the minimum time point and the length of the vehicle is less than the maximum value, a following decision is made for the obstacle; otherwise, an ignore strategy is adopted for the obstacle. The process of obtaining the speed curve for vehicle dynamic programming includes: By sampling on the ST graph, multiple feasible solutions are calculated in the forward direction over the entire time, and the minimum cost value of each feasible solution is obtained by reverse search. The speed curve corresponding to the minimum cost value is used as the speed curve for vehicle dynamic programming.

2. The vehicle speed decision method according to claim 1, characterized in that, The process of finding the target obstacle includes: Iterate through each obstacle that makes the follow decision to find the obstacle closest to the vehicle as the target obstacle.

3. The vehicle speed decision method according to claim 2, characterized in that, Also includes: If there is no obstacle that makes a follow decision, a virtual obstacle is constructed and used as the target obstacle, wherein the position of the virtual obstacle is at a distance greater than a preset distance from the vehicle and its speed is the reference speed of the vehicle.

4. The vehicle speed decision method according to claim 1, characterized in that, The step of performing a longitudinal simulation of the vehicle's forward direction within the lane based on the target obstacle to obtain the final speed curve includes: At each moment, calculate the expected distance between the vehicle and the target obstacle; Based on the desired distance, the acceleration of the improved intelligent driver model is calculated, and the acceleration of the iso-acceleration heuristic model is also calculated. The acceleration of the improved intelligent driver model and the acceleration of the iso-acceleration heuristic model are compared to obtain the acceleration at time t; Update the acceleration at time t for the next time step. The vehicle's position and the target obstacle's position are used to obtain the final speed curve based on the updated results.

5. A vehicle speed decision system, characterized in that, The system is used to implement the vehicle speed decision method as described in any one of claims 1-4, wherein the vehicle speed decision system includes: The acquisition module is used to obtain the speed curve of the vehicle's dynamic planning. The longitudinal decision-making module is used to make longitudinal decisions for each obstacle based on the speed curve, wherein the longitudinal decision-making includes a following strategy, an overtaking strategy, and an ignoring strategy. A search module for finding target obstacles, wherein the target obstacles include obstacles selected from those that have adopted a follow strategy; The speed decision module is used to perform longitudinal simulation of the vehicle's forward direction within the lane based on the target obstacle, and obtain the final speed curve.

6. A vehicle, characterized in that, include: The vehicle speed decision system according to claim 5.

7. A computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the vehicle speed decision method according to any one of claims 1-4.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the vehicle speed decision method according to any one of claims 1-4.

Images