Speed planning method and device, electronic equipment, vehicle and storage medium

Abstract

The invention provides a speed planning method and device, electronic equipment, a vehicle and a storage medium. The method comprises the steps that real-time sensing data of sensing moving objects around the vehicle by sensing equipment of the vehicle and real-time state data of the vehicle are acquired; wherein the moving object is an object moving relative to the ground; according to the real-time sensing data, the possibility that the moving object enters the front of the lane where the vehicle is located and interferes with the vehicle is predicted; and if it is predicted that the moving object has the possibility of interfering with the vehicle, according to the real-time sensing data and the real-time state data, planning a real-time vehicle speed at least keeping a safe distance between the vehicle and the moving object. According to the speed planning method provided by the invention, the real-time vehicle speed of the vehicle can be defensively planned on the premise of ensuring that the vehicle and the moving object at least keep the safe distance, so that poor driving and riding experience caused by sudden change of the real-time vehicle speed is avoided, and the driving and riding safety is ensured while the driving and riding comfort is improved.

Description

Technical Field

[0001] The present application relates to the field of intelligent driving technology, and in particular to a speed planning method, device, electronic device, vehicle and storage medium. Background Art

[0002] When a vehicle is driving on the road, especially when approaching complex sections such as intersections, it is often faced with situations where other vehicles in other lanes change lanes to the lane where the vehicle is located or pedestrians cross the road and interfere with the driving of the vehicle. In these situations, it is necessary to control the driving speed of the vehicle to ensure safety.

[0003] However, in the related art, when other vehicles or pedestrians interfere with the driving of the vehicle, it is usually necessary to control the vehicle to slow down immediately to avoid traffic accidents. The uneven change in driving speed will lead to poor driving and riding experience of the vehicle.

[0004] Therefore, it is necessary to provide a technical solution to solve the above technical problems. Summary of the invention

[0005] The present application provides a speed planning method, device, electronic device, vehicle and storage medium, which can plan the speed of a vehicle when it is predicted that surrounding moving objects may interfere with the vehicle.

[0006] In a first aspect, the present application provides a speed planning method, applied to a vehicle, comprising:

[0007] Acquire real-time perception data of moving objects around the vehicle and real-time status data of the vehicle, which are perceived by the vehicle's perception device; wherein the moving objects are objects moving relative to the ground;

[0008] Predicting, based on the real-time perception data, the possibility that the moving object enters in front of the lane where the ego vehicle is located and interferes with the ego vehicle;

[0009] If it is predicted that the moving object has the possibility of interfering with the ego vehicle, a real-time vehicle speed is planned so that the ego vehicle maintains at least a safe distance from the moving object based on the real-time perception data and the real-time status data.

[0010] Optionally, predicting, based on the real-time perception data, the possibility that the moving object enters in front of the lane where the ego vehicle is located and interferes with the ego vehicle includes:

[0011] Predicting a movement path of the mobile object according to the real-time sensing data, and calculating a probability that the mobile object enters the predicted movement path;

[0012] If it is predicted that the moving object has the possibility of interfering with the ego vehicle, then according to the real-time perception data and the real-time status data, planning a real-time speed of the ego vehicle so as to keep at least a safe distance from the moving object includes:

[0013] If the predicted movement route includes a route that enters the lane where the ego vehicle is located, determining the interference caused by the moving object on the ego vehicle according to the probability that the moving object enters the predicted movement route;

[0014] A current recommended speed of the vehicle is determined based on the real-time perception data and the real-time status data so that interference of the moving object on the vehicle is minimized.

[0015] Optionally, if the predicted movement route includes a route entering the lane where the ego vehicle is located, determining the interference caused by the mobile object on the ego vehicle according to the probability of the mobile object entering the predicted movement route includes:

[0016] According to the probability of the mobile object entering the predicted motion route, constructing a loss function of the vehicle; wherein the loss function includes a safety loss function and a comfort loss function;

[0017] The determining, based on the real-time perception data and the real-time status data, the current recommended speed of the vehicle when the interference of the moving object on the vehicle is minimized includes:

[0018] According to the real-time perception data and the real-time status data, an ST graph is established with a reference line of the lane where the vehicle is located as an S axis and time as a T axis, and the predicted movement route is projected onto the ST graph;

[0019] The loss function is minimized and evaluated based on the ST graph to obtain the current recommended speed of the vehicle when the value of the loss function is minimized.

[0020] Optionally, after minimizing the loss function based on the ST graph to obtain the current recommended speed of the vehicle when the value of the loss function is minimized, the speed planning method further includes:

[0021] According to the real-time perception data and the real-time status data, at a preset refresh interval, a ST map is re-established with a reference line of the lane where the vehicle is located as an S axis and time as a T axis, and the predicted movement route is re-projected onto the ST map;

[0022] The loss function is minimized and evaluated based on the re-established ST graph to obtain a recommended speed of the vehicle at the refresh interval when the value of the loss function is minimized.

[0023] Optionally, the loss function is related to the motion state of the vehicle; and the minimizing evaluation of the loss function based on the ST graph includes:

[0024] According to the preset value interval, a preset number of time points are selected starting from zero;

[0025] Substituting the selected time point into the ST diagram to obtain the value of the physical quantity of the vehicle at the selected time point regarding the motion state;

[0026] Substitute the value of the physical quantity of the vehicle regarding the motion state at the selected time point into the loss function to obtain the value of the loss function.

[0027] Optionally, at least one of the predicted movement routes;

[0028] The constructing the loss function of the vehicle according to the probability that the moving object enters the predicted motion path includes:

[0029] Constructing a sub-loss function of the ego vehicle when the mobile object enters each predicted motion path;

[0030] The sum of the products of the probability of the mobile object entering each predicted motion path and the corresponding sub-loss function is used as the loss function of the vehicle.

[0031] Optionally, the loss function also includes a traffic efficiency loss function;

[0032] The constructing of the sub-loss function of the vehicle when the mobile object enters each predicted motion path includes:

[0033] When the mobile object enters each predicted motion route, the sum of the traffic efficiency loss function, the comfort loss function and the safety loss function of the ego vehicle is used as the sub-loss function of the ego vehicle.

[0034] Optionally, the traffic efficiency loss function is constructed in the following manner:

[0035] According to the real-time status data, obtaining the real-time speed and real-time position of the vehicle;

[0036] Determining a first difference between the real-time speed and a preset expected speed and a second difference between the real-time position and a preset expected position;

[0037] Constructing the traffic efficiency loss function according to the first difference and the second difference and a preset speed weight coefficient and a position weight coefficient;

[0038] The comfort loss function is constructed as follows:

[0039] According to the real-time status data, obtaining the real-time acceleration and real-time jerk of the vehicle;

[0040] Constructing the comfort loss function according to the real-time acceleration and the real-time jerk and the preset acceleration weight coefficient and jerk weight coefficient;

[0041] The security loss function is constructed as follows:

[0042] Acquire the real-time position of the vehicle and the real-time distance between the vehicle and the moving object according to the real-time perception data and the real-time status data;

[0043] determining a third difference between the real-time distance and the safety distance;

[0044] The safety loss function is constructed based on the real-time position, the third difference and a preset safety weight coefficient.

[0045] Optionally, if it is predicted that the moving object has the possibility of interfering with the ego vehicle, planning a real-time vehicle speed at which the ego vehicle maintains at least a safe distance from the moving object according to the real-time perception data and the real-time status data includes:

[0046] If it is predicted that the moving object has the possibility of interfering with the vehicle, judging whether the vehicle is currently traveling on a road section where overtaking is not allowed based on the real-time status data;

[0047] If yes, planning a real-time speed of the ego vehicle to maintain at least a safe distance behind the moving object according to the real-time perception data and the real-time status data;

[0048] If not, the real-time vehicle speed of the vehicle is planned to maintain at least a safe distance in front of the moving object according to the real-time perception data and the real-time status data.

[0049] In a second aspect, the present application provides a speed planning device, applied to a vehicle, comprising:

[0050] An acquisition module, used to acquire real-time perception data of a vehicle's perception device on a moving object around the vehicle and real-time status data of the vehicle; wherein the moving object is an object moving relative to the ground;

[0051] A prediction module, used for predicting the possibility that the moving object enters the front of the lane where the ego vehicle is located and interferes with the ego vehicle based on the real-time perception data;

[0052] The planning module is used to plan the real-time speed of the ego vehicle to maintain at least a safe distance from the moving object based on the real-time perception data and the real-time status data if it is predicted that the moving object has the possibility of interfering with the ego vehicle.

[0053] In a third aspect, the present application provides an electronic device comprising at least one processor, wherein the processor is used to implement the speed planning method as described in any one of the first aspects.

[0054] In a fourth aspect, the present application provides a vehicle, comprising at least one processor, wherein the processor is used to implement the speed planning method as described in any one of the first aspects.

[0055] In a fifth aspect, the present application provides a computer-readable storage medium having a program stored thereon, which, when executed by a processor, implements the speed planning method as described in any one of the first aspects.

[0056] In the speed planning method, device, electronic device, vehicle and storage medium provided by the present application, the possibility of the moving object entering the lane is predicted based on the real-time perception data of the moving objects around the vehicle. When it is predicted that the moving object may interfere with the vehicle, the real-time speed of the vehicle is planned. The real-time speed of the vehicle can be defensively planned before the moving object interferes with the vehicle, avoiding the poor driving and riding experience of the vehicle caused by the sudden change of the real-time speed. In addition, when planning the real-time speed of the vehicle, it is also ensured that the vehicle and the moving object maintain at least a safe distance, which improves the comfort of driving and riding the vehicle while ensuring the safety of driving and riding the vehicle. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

[0058] Figure 1 Shown is a flow chart of a speed planning method according to an embodiment of the present application.

[0059] Figure 2 Shown is a schematic diagram of a usage scenario of the speed planning method according to an embodiment of the present application.

[0060] Figure 3 The ST diagram created in the speed planning method according to an embodiment of the present application is shown.

[0061] Figure 4Shown is a structural block diagram of a speed planning device according to an embodiment of the present application.

[0062] Figure 5 Shown is a structural block diagram of an electronic device according to an embodiment of the present application. DETAILED DESCRIPTION

[0063] Here, the technical solutions in the embodiments of the present application will be described clearly and completely in conjunction with the accompanying drawings. When the following description refers to the accompanying drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements.

[0064] The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with one or more embodiments of this specification. Instead, they are merely examples of devices and methods consistent with some aspects of one or more embodiments of this specification as detailed in the attached claims. It should be noted that in other embodiments, the steps of the corresponding method are not necessarily performed in the order shown and described in this specification. In some other embodiments, the steps included in the method may be more or less than those described in this specification. In addition, a single step described in this specification may be decomposed into multiple steps for description in other embodiments; and multiple steps described in this specification may be combined into a single step for description in other embodiments.

[0065] In order to solve the technical problem in the related art that when the moving objects around the vehicle interfere with the driving of the vehicle, the real-time speed of the vehicle changes suddenly, which affects the driving and riding experience of the vehicle, the embodiment of the present application provides a speed planning method, which predicts the possibility of the moving object entering the lane based on the real-time perception data of the moving objects around the vehicle, and plans the real-time speed of the vehicle when it is predicted that the moving object may interfere with the vehicle. The real-time speed of the vehicle can be defensively planned before the moving object interferes with the vehicle, avoiding the poor driving and riding experience of the vehicle caused by the sudden change of the real-time speed. In addition, when planning the real-time speed of the vehicle, it is also ensured that the vehicle and the moving object maintain at least a safe distance, which improves the driving and riding comfort of the vehicle while ensuring the driving and riding safety of the vehicle.

[0066] See also Figure 1 and Figure 2 , Figure 1 FIG. 1 is a flow chart of a method for speed planning according to an embodiment of the present application. Figure 2 The speed planning method provided in the embodiment of the present application includes but is not limited to the following steps S101 to S103.

[0067] S101, obtaining real-time perception data of moving objects around the vehicle and real-time status data of the vehicle through the perception device of the vehicle; wherein the moving objects are objects moving relative to the ground.

[0068] The vehicle's sensing device may be a variety of sensors installed on the vehicle, such as millimeter wave radar, laser radar, single / binocular camera, and satellite navigation. Through the vehicle's sensing device, the vehicle's surrounding environment can be sensed in real time during the vehicle's driving process, and real-time sensing data for sensing the vehicle's surrounding environment can be obtained. Based on the real-time sensing data, static and dynamic objects can be identified, detected, and tracked, so as to determine which sensing objects in the vehicle's surrounding environment are moving objects that move relative to the ground, such as motor vehicles, non-motor vehicles, pedestrians, etc., and then real-time sensing data for sensing moving objects around the vehicle can be obtained from the real-time sensing data for sensing the vehicle's surrounding environment.

[0069] In addition, when acquiring real-time perception data of moving objects around the vehicle, the real-time status data of the vehicle can also be acquired, such as remaining power, estimated mileage, real-time location, real-time vehicle speed, real-time acceleration, etc., so as to grasp the real-time status of the vehicle.

[0070] S102: predicting, based on the real-time perception data, the possibility that the moving object enters the front of the lane where the ego vehicle is located and interferes with the ego vehicle.

[0071] Based on the real-time perception data of the moving objects around the vehicle, the movement path of the moving objects can be predicted. Figure 2 In the scenario shown: since his car 1 is still in the right lane when it reaches the intersection, his car 1 will not turn left at the intersection; since his car 1 turning right at the intersection will not interfere with the driving of his own car, the situation where his car 1 turns right at the intersection is not considered; therefore, his car 1 may only have two movement routes, Route 1 and Route 2; since his car 2 is still in the right lane when it reaches the intersection, his car 2 will not turn left at the intersection; since the horizontal and longitudinal lanes will not allow passage at the same time, the situation where his car 2 goes straight at the intersection is not considered; therefore, his car 2 may only have two movement routes, Route 3 and Route 4.

[0072] Among them, because the other car 1 will interfere with the driving of the own car before the other car 2, we can temporarily only consider the interference caused by the other car 1 on the driving of the own car. For example, if it is sensed that the other car 1 has turned on the left turn signal or the body of the other car 1 is sensed to be tilted to the left, it can be predicted that the other car 1 may enter the route 1, that is, the other car 1 may change lanes to the left to the lane where the own car is located. Because the route 1 and the driving route of the own car intersect and overlap, the other car 1 may enter the front of the lane where the own car is located and interfere with the driving of the own car.

[0073] S103, if it is predicted that the moving object has the possibility of interfering with the ego vehicle, then based on the real-time perception data and the real-time status data, plan a real-time speed at which the ego vehicle maintains at least a safe distance from the moving object.

[0074] When it is predicted that a moving object may interfere with the ego vehicle, the real-time speed of the ego vehicle is planned before the moving object enters the lane in front of the ego vehicle and interferes with the ego vehicle. This can make the real-time speed of the ego vehicle continue to change smoothly and avoid the discomfort caused by sudden and drastic changes in the real-time speed.

[0075] Specifically, the real-time status of the moving object can be obtained based on the real-time perception data of the moving object, and the real-time status of the own vehicle can be obtained based on the real-time status data of the own vehicle, so as to plan the real-time vehicle speed that keeps at least a safe distance from the moving object based on the real-time status of the two.

[0076] In some embodiments:

[0077] In the above step S102, when predicting the possibility of the moving object entering the front of the lane where the vehicle is located and interfering with the vehicle based on the real-time perception data, the following steps are specifically included:

[0078] Predicting the movement path of the mobile object according to the real-time sensing data, and calculating the probability that the mobile object enters the predicted movement path;

[0079] In the above step S103, if it is predicted that the moving object has the possibility of interfering with the ego vehicle, then according to the real-time perception data and the real-time status data, planning the real-time speed of the ego vehicle to keep at least a safe distance from the moving object specifically includes the following steps:

[0080] Step 1: If the predicted movement path includes a path that enters the lane where the ego vehicle is located, then the interference caused by the moving object on the ego vehicle is determined according to the probability of the moving object entering the predicted movement path;

[0081] Step 2: Determine the current recommended speed of the vehicle under the condition that the interference of the moving object to the vehicle is minimized according to the real-time perception data and the real-time status data.

[0082] If the predicted movement route of the mobile object includes a route that enters the lane where the ego vehicle is located, it means that the predicted movement route of the mobile object and the driving route of the ego vehicle intersect or overlap, that is, the mobile object may interfere with the ego vehicle. At this time, the interference of the mobile object on the ego vehicle can be determined based on the probability of the mobile object entering the predicted movement route, and the recommended speed of the ego vehicle under the condition that the interference of the mobile object on the ego vehicle is minimized can be determined based on the real-time perception data of the mobile object and the real-time status data of the ego vehicle.

[0083] In some embodiments:

[0084] In the above steps, if the predicted movement route includes a route that enters the lane where the ego vehicle is located, then according to the probability of the mobile object entering the predicted movement route, when determining the interference of the mobile object on the ego vehicle, the following steps are specifically included:

[0085] According to the probability of the moving object entering the predicted motion route, a loss function of the vehicle is constructed; wherein the loss function includes a safety loss function and a comfort loss function;

[0086] In the above steps, when determining the current recommended speed of the vehicle under the condition that the interference of the moving object to the vehicle is minimized according to the real-time perception data and the real-time status data, the following steps are specifically included:

[0087] Step 1: Based on the real-time perception data and the real-time status data, an ST graph is established with a reference line of the lane where the vehicle is located as an S axis and time as a T axis, and the predicted movement route is projected onto the ST graph;

[0088] Step 2: Minimize the loss function based on the ST graph to obtain the current recommended speed of the vehicle when the loss function is minimized.

[0089] The interference caused by the moving object on the ego vehicle is quantified, which is the loss function of the ego vehicle. The moving object interferes with the driving of the ego vehicle, and the moving object affects the driving and riding safety of the ego vehicle; avoiding the moving object requires changing the real-time speed of the ego vehicle, and the real-time speed change will affect the driving and riding comfort of the ego vehicle. Therefore, the loss function of the ego vehicle includes a safety loss function that represents the impact on safety and a comfort loss function that represents the impact on comfort.

[0090] Determine the recommended speed of the vehicle when the interference of the moving object on the vehicle is minimized, that is, obtain the recommended speed of the vehicle when the value of the loss function of the vehicle is minimized.

[0091] Please also read Figures 1 to 3 , Figure 3The ST diagram established in the speed planning method of the embodiment of the present application is shown. The ST diagram is established with the reference line of the lane where the vehicle is located as the S axis and time as the T axis. The slope of the curve at a certain point in the ST diagram is the real-time speed of the vehicle at the corresponding time at that point.

[0092] If the other vehicle 1 enters route 1, then from the moment (t-0) when the other vehicle 1 changes lanes to the lane where the vehicle is located, the movement route of the other vehicle 1 will continue to exist in the ST graph. Projecting route 1 into the ST graph is Figure 3 The middle gray parallelogram. In order to avoid the other car 1, the curve corresponding to the speed sequence 1 of the ego vehicle cannot intersect with the gray parallelogram; because overtaking is not allowed at the intersection, the ego vehicle can only drive behind the other car 1, so the curve corresponding to the speed sequence 1 of the ego vehicle can only be located below the gray parallelogram. If the other car 1 enters route 2, the movement path of the other car 1 will not exist in the ST diagram, that is, there is no projection of route 2 in the ST diagram. Because the other car 1 on route 2 will not interfere with the ego vehicle, the ego vehicle can be controlled to accelerate after slowing down to pass the intersection.

[0093] In some embodiments, in the above steps, the loss function is minimized based on the ST graph to obtain the minimum value of the loss function, and after the current recommended speed of the vehicle, the speed planning method further includes:

[0094] Step 1: at a preset refresh interval, based on the real-time perception data and the real-time status data, re-establishing the ST map with the reference line of the lane where the vehicle is located as the S axis and time as the T axis, and re-projecting the predicted movement route onto the ST map;

[0095] Step 2: Minimize the loss function based on the re-established ST graph to obtain the recommended speed of the vehicle at the refresh interval when the value of the loss function is minimized.

[0096] The currently established ST graph is a real-time state prediction of the ego vehicle and the mobile object from 0s to 5s from the current time. Based on the currently established ST graph, the loss function of the ego vehicle is minimized and evaluated, and the recommended speed of the ego vehicle that minimizes the value of the loss function is obtained. However, the real-time state of the ego vehicle and the mobile object is constantly changing. Therefore, it is necessary to re-establish the ST graph according to a preset refresh interval, such as 0.1s, and obtain the recommended speed of the ego vehicle that minimizes the value of the loss function at the refresh interval based on the ST graph re-established at the refresh interval. By analogy, the recommended speed of the ego vehicle at each refresh interval is obtained. Controlling the ego vehicle to travel at the recommended speed can minimize the value of the loss function of the ego vehicle at each moment, that is, to minimize the interference caused by the mobile object entering the lane in front of the ego vehicle.

[0097] In some embodiments, the loss function in the above steps is related to the motion state of the ego vehicle;

[0098] In the above steps, when minimizing the loss function based on the ST graph, the following steps are specifically included:

[0099] Step 1, selecting a preset number of time points starting from zero according to a preset value interval;

[0100] Step 2: Substitute the selected time point into the ST diagram to obtain the value of the physical quantity of the vehicle's motion state at the selected time point;

[0101] Step three, substitute the value of the physical quantity of the vehicle's motion state at the selected time point into the loss function to obtain the value of the loss function.

[0102] The physical quantity of the vehicle regarding the motion state at the selected time point may include multiple ones of real-time position, real-time speed, real-time acceleration, real-time jerk, and real-time distance from the moving object.

[0103] In some embodiments, at least one of the predicted motion routes in the above steps;

[0104] In the above steps, according to the probability of the mobile object entering the predicted movement path, the loss function of the vehicle is constructed, which specifically includes the following steps:

[0105] Step 1: construct a sub-loss function of the vehicle when the mobile object enters each predicted motion path;

[0106] Step 2: The sum of the product of the probability of the mobile object entering each predicted motion path and its corresponding sub-loss function is used as the loss function of the vehicle.

[0107] Because the moving object entering different motion routes will cause different interferences to the ego vehicle, it is necessary to construct a sub-loss function of the ego vehicle when the moving object enters each predicted motion route. Because the probability of the moving object entering different motion routes is different, the weight of the sub-loss function corresponding to the moving object entering each predicted motion route is different. Therefore, the probability of the moving object entering each predicted motion route is used as the weight coefficient, and the expression of the weighted average of all sub-loss functions is the loss function of the ego vehicle. By using the probability of the moving object entering each predicted motion route as the weight coefficient of its corresponding sub-loss function, the different interferences caused by the moving object entering motion routes with different probabilities on the ego vehicle can be comprehensively considered, making the loss function of the ego vehicle more accurate.

[0108] The above method of constructing the loss function of the self-driving car can be summarized as the following formula:

[0109] f total =p1×f1+p2×f2+…+p m ×f m

[0110] Among them, f total is the loss function of the vehicle, f m is the sub-loss function of the vehicle when the mobile object enters each predicted motion path, p m is the probability of the mobile object entering each predicted motion path, and m is the number of predicted motion paths.

[0111] In some embodiments, the loss function in the above steps further includes a traffic efficiency loss function;

[0112] In the above steps, when constructing the sub-loss function of the vehicle when the mobile object enters each predicted motion path, the specific steps include:

[0113] When the mobile object enters each predicted motion route, the sum of the vehicle's traffic efficiency loss function, comfort loss function, and safety loss function is used as the vehicle's sub-loss function.

[0114] In order to avoid the moving object, the real-time speed of the vehicle needs to be reduced. However, reducing the real-time speed of the vehicle will inevitably reduce the traffic efficiency of the vehicle. Therefore, the loss function of the vehicle also includes the traffic efficiency loss function. When the moving object enters each predicted movement route, the sum of the safety loss function, comfort loss function and traffic efficiency loss function of the vehicle is the sub-loss function of the vehicle when the moving object enters each predicted movement route. By adding the safety loss function, comfort loss function and traffic efficiency loss function of the vehicle as the sub-loss function of the vehicle when the moving object enters each predicted movement route, the losses in the three aspects of safety, comfort and traffic efficiency can be comprehensively considered when evaluating the loss function of the vehicle, making the loss function of the vehicle more comprehensive.

[0115] The above method of constructing the sub-loss function of the vehicle when the mobile object enters each predicted motion path can be summarized as the following formula:

[0116] f m =f safe +f comfort +f efficiency

[0117] Among them, f safe 、f comfort 、f efficiency They are respectively the safety loss function of the vehicle, the comfort loss function of the vehicle, and the traffic efficiency loss function of the vehicle.

[0118] In some embodiments, the traffic efficiency loss function in the above steps is constructed as follows:

[0119] Step 1: Obtain the real-time speed and real-time position of the vehicle according to the real-time status data;

[0120] Step 2, determining a first difference between the real-time speed and a preset expected speed and a second difference between the real-time position and a preset expected position;

[0121] Step three, constructing the traffic efficiency loss function according to the first difference, the second difference, and the preset speed weight coefficient and position weight coefficient.

[0122] The above method of constructing the traffic efficiency loss function can be summarized as the following formula:

[0123]

[0124] Among them, v i and i are the real-time speed and real-time position of the vehicle, v desire and desire are the preset expected speed and expected position, w v and w s are the preset speed weight coefficient and position weight coefficient respectively, and i is an integer. By selecting a preset number of time points starting from zero according to a preset value interval and substituting these time points into the ST diagram, the real-time position and real-time speed of the vehicle at these time points can be obtained.

[0125] For example, if the preset value interval is 0.2, then 26 time points can be obtained from 0s to 5s, and the preset number n is 26. Substituting these 26 time points into the ST graph, the real-time positions s0, s1, s2, ..., s5 of the vehicle at 0s, 0.2s, 0.4s, ..., 4.8s, and 5s can be obtained respectively. 24 、s 25 and real-time speed v0, v1, v2, ..., v 24 、v 25 .

[0126] In some embodiments, the comfort loss function in the above steps is constructed as follows:

[0127] Step 1: obtaining the real-time acceleration and real-time jerk of the vehicle according to the real-time status data;

[0128] Step 2: construct the comfort loss function according to the real-time acceleration and the real-time jerk and the preset acceleration weight coefficient and jerk weight coefficient.

[0129] The above method of constructing the comfort loss function can be summarized as the following formula:

[0130]

[0131] Among them, a i and j i are the real-time acceleration and real-time jerk of the vehicle, respectively, w a and w j are respectively the preset acceleration weight coefficient and jerk weight coefficient, and i is an integer. By selecting a preset number of time points starting from zero according to a preset value interval, and substituting these time points into the ST diagram, the real-time acceleration and real-time jerk of the vehicle at these time points can be obtained.

[0132] For example, if the preset value interval is 0.2, then 26 time points can be taken from 0s to 5s, and the preset number n is 26. Substituting these 26 time points into the ST diagram, the real-time acceleration a0, a1, a2, ..., a3 of the vehicle at 0s, 0.2s, 0.4s, ..., 4.8s, and 5s can be obtained respectively. 24 、a 25 and real-time speed j0, j1, j2, ..., j 24 、j 25 .

[0133] In some embodiments, the security loss function in the above steps is constructed as follows:

[0134] Step 1: acquiring the real-time position of the vehicle and the real-time distance between the vehicle and the moving object according to the real-time sensing data and the real-time status data;

[0135] Step 2, determining a third difference between the real-time distance and the safety distance;

[0136] Step three, constructing the security loss function according to the real-time position, the third difference and a preset security weight coefficient.

[0137] The above method of constructing the security loss function can be summarized as the following formula:

[0138]

[0139] Among them, s i is the real-time position of the vehicle, s obs is the real-time distance between the vehicle and the moving object, d safe is the safe distance between the vehicle and the moving object, w safeis the preset safety weight coefficient, i is an integer. By selecting a preset number of time points from zero according to a preset value interval, and substituting these time points into the ST diagram, the real-time position and real-time speed of the vehicle at these time points can be obtained.

[0140] For example, if the preset value interval is 0.2, then 26 time points can be obtained from 0s to 5s, and the preset number n is 26. Substituting these 26 time points into the ST graph, the real-time positions s0, s1, s2, ..., s5 of the vehicle at 0s, 0.2s, 0.4s, ..., 4.8s, and 5s can be obtained respectively. 24 、s 25 .

[0141] In some embodiments, in the above step S103, if it is predicted that the moving object has the possibility of interfering with the ego vehicle, then according to the real-time perception data and the real-time status data, planning the real-time speed of the ego vehicle to maintain at least a safe distance from the moving object specifically includes the following steps:

[0142] Step 1: If it is predicted that the moving object may interfere with the vehicle, it is determined whether the vehicle is currently traveling on a road section where overtaking is not allowed based on the real-time status data;

[0143] Step 2: If yes, then plan the real-time speed of the vehicle to keep at least a safe distance behind the moving object based on the real-time perception data and the real-time status data;

[0144] Step three: if not, then based on the real-time perception data and the real-time status data, plan the real-time speed of the vehicle to maintain at least a safe distance in front of the moving object.

[0145] If it is predicted that a moving object may enter the front of the lane where the vehicle is located and interfere with the vehicle, the speed of the vehicle needs to be controlled to avoid the moving object. In some complex sections, such as intersections, tunnels, ramps, etc., vehicles are not allowed to overtake. The vehicle can be controlled to slow down so as to keep a distance from the moving object. The real-time speed of the vehicle can be planned to keep at least a safe distance behind the moving object. In some ordinary sections, vehicles are allowed to overtake. The vehicle can be controlled to accelerate so as to surpass the moving object before the moving object enters the lane where the vehicle is located. The real-time speed of the vehicle can be planned to keep at least a safe distance in front of the moving object. This not only ensures the safety and comfort of driving and riding the vehicle, but also improves the driving and riding efficiency of the vehicle.

[0146] The present application also provides a speed planning device, see Figure 4 , Figure 4The structure diagram of the speed planning device of the embodiment of the present application is shown. The speed planning device 21 provided in the embodiment of the present application may include:

[0147] The acquisition module 211 is used to acquire real-time perception data of the vehicle's perception device on the moving objects around the vehicle and the real-time status data of the vehicle; wherein the moving objects are objects moving relative to the ground;

[0148] A prediction module 212, configured to predict, based on the real-time perception data, the possibility that the moving object enters the front of the lane where the ego vehicle is located and interferes with the ego vehicle;

[0149] The planning module 213 is used to plan the real-time vehicle speed so that the vehicle and the moving object maintain at least a safe distance based on the real-time perception data and the real-time status data if it is predicted that the moving object has the possibility of interfering with the vehicle.

[0150] The implementation process of the functions and effects of each module in the above-mentioned device is specifically described in the implementation process of the corresponding steps in the above-mentioned method, and the same technical effect can be achieved, which will not be repeated here.

[0151] In some embodiments:

[0152] When predicting the possibility that the mobile object enters the front of the lane where the vehicle is located and interferes with the vehicle based on the real-time perception data, the prediction module 212 is specifically used to: predict the movement path of the mobile object based on the real-time perception data, and calculate the probability that the mobile object enters the predicted movement path;

[0153] If it is predicted that the moving object has the possibility of interfering with the ego vehicle, the planning module 213 plans the real-time vehicle speed of the ego vehicle to maintain at least a safe distance from the moving object based on the real-time perception data and the real-time status data. Specifically, it is used to: if the predicted movement route includes a route entering the lane where the ego vehicle is located, determine the interference caused by the moving object on the ego vehicle based on the probability of the moving object entering the predicted movement route; and determine the current recommended speed of the ego vehicle under the condition that the interference caused by the moving object on the ego vehicle is minimized based on the real-time perception data and the real-time status data.

[0154] In some embodiments:

[0155] When the predicted movement route includes a route that enters the lane where the ego vehicle is located, the planning module 213 determines the interference of the mobile object on the ego vehicle according to the probability of the mobile object entering the predicted movement route, and is specifically used to: construct a loss function of the ego vehicle according to the probability of the mobile object entering the predicted movement route; wherein the loss function includes a safety loss function and a comfort loss function;

[0156] When the planning module 213 determines the current recommended speed of the vehicle under the condition that the interference of the moving object on the vehicle is minimized based on the real-time perception data and the real-time status data, it is specifically used to: establish an ST diagram based on the real-time perception data and the real-time status data, with the reference line of the lane where the vehicle is located as the S axis and time as the T axis, and project the predicted movement route onto the ST diagram; minimize the loss function based on the ST diagram to obtain the current recommended speed of the vehicle under the condition that the value of the loss function is minimized.

[0157] In some embodiments, after minimizing the loss function based on the ST diagram and obtaining the current recommended speed of the vehicle when the value of the loss function is minimized, the planning module 213 is also used to: re-establish the ST diagram according to the preset refresh interval, based on the real-time perception data and the real-time status data, with the reference line of the lane where the vehicle is located as the S axis and time as the T axis, and re-project the predicted movement route onto the ST diagram; minimize the loss function based on the re-established ST diagram and obtain the recommended speed of the vehicle at the refresh interval when the value of the loss function is minimized.

[0158] In some embodiments, the above loss function is related to the motion state of the ego vehicle;

[0159] When the planning module 213 performs minimization evaluation on the loss function based on the ST diagram, it is specifically used to: select a preset number of time points starting from zero according to a preset value interval; substitute the selected time points into the ST diagram to obtain the values ​​of physical quantities related to the motion state of the vehicle at the selected time points; substitute the values ​​of physical quantities related to the motion state of the vehicle at the selected time points into the loss function to obtain the value of the loss function.

[0160] In some embodiments, at least one of the predicted movement routes described above;

[0161] When constructing the loss function of the vehicle according to the probability of the mobile object entering the predicted motion route, the planning module 213 is specifically used to: construct a sub-loss function of the vehicle when the mobile object enters each predicted motion route; and take the sum of the products of the probability of the mobile object entering each predicted motion route and its corresponding sub-loss function as the loss function of the vehicle.

[0162] In some embodiments, the above-mentioned loss function also includes a traffic efficiency loss function;

[0163] When constructing the sub-loss function of the vehicle when the mobile object enters each predicted motion route, the planning module 213 is specifically used to: when the mobile object enters each predicted motion route, the sum of the vehicle's traffic efficiency loss function, comfort loss function and safety loss function is used as the sub-loss function of the vehicle.

[0164] In some embodiments, when constructing the traffic efficiency loss function of the vehicle, the planning module 213 is specifically used to: obtain the real-time speed and real-time position of the vehicle based on the real-time status data; determine a first difference between the real-time speed and a preset expected speed and a second difference between the real-time position and the preset expected position; and construct the traffic efficiency loss function based on the first difference and the second difference and the preset speed weight coefficient and position weight coefficient.

[0165] In some embodiments, when constructing the comfort loss function of the vehicle, the planning module 213 is specifically used to: obtain the real-time acceleration and real-time jerk of the vehicle according to the real-time status data; and construct the comfort loss function according to the real-time acceleration and the real-time jerk as well as the preset acceleration weight coefficient and the jerk weight coefficient.

[0166] In some embodiments, when constructing the safety loss function of the vehicle, the planning module 213 is specifically used to: obtain the real-time position of the vehicle and the real-time distance between the vehicle and the moving object based on the real-time perception data and the real-time status data; determine the third difference between the real-time distance and the safety distance; and construct the safety loss function based on the real-time position and the third difference and a preset safety weight coefficient.

[0167] In some embodiments, if it is predicted that the moving object may interfere with the own vehicle, the planning module 213 plans the real-time speed of the own vehicle to maintain at least a safe distance from the moving object based on the real-time perception data and the real-time status data. Specifically, it is used to: if it is predicted that the moving object may interfere with the own vehicle, determine whether the own vehicle is currently traveling on a road section where overtaking is not allowed based on the real-time status data; if so, plan the real-time speed of the own vehicle to maintain at least a safe distance behind the moving object based on the real-time perception data and the real-time status data; if not, plan the real-time speed of the own vehicle to maintain at least a safe distance in front of the moving object based on the real-time perception data and the real-time status data.

[0168] The present application also provides an electronic device, which may include the above-mentioned speed planning device 21. Figure 5 , Figure 5 The electronic device 20 may include one or more processors 22, and the processor 22 is used to implement the above-mentioned speed planning method.

[0169] In some embodiments, the electronic device 20 may include a computer-readable storage medium 23, which may store a program that can be called by the processor 22 and may include a non-volatile storage medium. In other embodiments, the electronic device 20 may also include a memory 24 and an interface 25. In other embodiments, the electronic device 20 may also include other hardware according to actual applications.

[0170] The computer-readable storage medium 23 provided in the embodiment of the present application stores a program thereon, and when the program is executed by the processor 22, it is used to implement the above-mentioned speed planning method.

[0171] The present application may take the form of a computer program product implemented on one or more computer-readable storage media 23 (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing program code. The computer-readable storage medium 23 includes permanent and non-permanent, removable and non-removable media, and information storage can be implemented by any method or technology. The information can be a computer-readable instruction, a data structure, a module of a program, or other data. The computer-readable storage medium 23 includes but is not limited to: phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, read-only compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassette, tape disk storage or other magnetic storage device or any other non-transmission medium, which can be used to store information that can be accessed by a computing device.

[0172] The above description is only a preferred embodiment of this specification and is not intended to limit this specification. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of this specification should be included in the scope of protection of this specification.

[0173] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the phrase "includes a ..." defines an element, and does not exclude the presence of other identical elements in the process, method, commodity or device including the element.

Claims

1. A speed planning method, applied to a vehicle, characterized in that: include: Acquire real-time perception data of moving objects around the vehicle and real-time status data of the vehicle, which are perceived by the vehicle's perception device; wherein the moving objects are objects moving relative to the ground; Predicting, based on the real-time perception data, the possibility that the moving object enters in front of the lane where the ego vehicle is located and interferes with the ego vehicle; If it is predicted that the moving object has the possibility of interfering with the ego vehicle, a real-time vehicle speed is planned so that the ego vehicle maintains at least a safe distance from the moving object based on the real-time perception data and the real-time status data.

2. The speed planning method according to claim 1, characterized in that: The predicting, based on the real-time perception data, the possibility that the moving object enters the front of the lane where the ego vehicle is located and interferes with the ego vehicle includes: Predicting a movement path of the mobile object according to the real-time sensing data, and calculating a probability that the mobile object enters the predicted movement path; If it is predicted that the moving object has the possibility of interfering with the ego vehicle, then according to the real-time perception data and the real-time status data, planning a real-time speed of the ego vehicle so as to keep at least a safe distance from the moving object includes: If the predicted movement route includes a route that enters the lane where the ego vehicle is located, determining the interference caused by the moving object on the ego vehicle according to the probability that the moving object enters the predicted movement route; A current recommended speed of the vehicle is determined based on the real-time perception data and the real-time status data so that interference of the moving object on the vehicle is minimized.

3. The speed planning method according to claim 2, characterized in that: If the predicted movement route includes a route entering the lane where the ego vehicle is located, determining the interference of the mobile object on the ego vehicle according to the probability of the mobile object entering the predicted movement route, including: According to the probability of the mobile object entering the predicted motion route, constructing a loss function of the vehicle; wherein the loss function includes a safety loss function and a comfort loss function; The determining, based on the real-time perception data and the real-time status data, the current recommended speed of the vehicle when the interference of the moving object on the vehicle is minimized includes: According to the real-time perception data and the real-time status data, an ST graph is established with a reference line of the lane where the vehicle is located as an S axis and time as a T axis, and the predicted movement route is projected onto the ST graph; The loss function is minimized and evaluated based on the ST graph to obtain the current recommended speed of the vehicle when the value of the loss function is minimized.

4. The speed planning method according to claim 3, characterized in that: After minimizing the loss function based on the ST graph to obtain the current recommended speed of the vehicle when the value of the loss function is minimized, the speed planning method further includes: According to the real-time perception data and the real-time status data, at a preset refresh interval, a ST map is re-established with a reference line of the lane where the vehicle is located as an S axis and time as a T axis, and the predicted movement route is re-projected onto the ST map; The loss function is minimized and evaluated based on the re-established ST graph to obtain a recommended speed of the vehicle at the refresh interval when the value of the loss function is minimized.

5. The method for speed planning according to claim 3, characterized in that: The loss function is related to the motion state of the vehicle; and the minimization evaluation of the loss function based on the ST graph includes: According to the preset value interval, a preset number of time points are selected starting from zero; Substituting the selected time point into the ST diagram to obtain the value of the physical quantity of the vehicle at the selected time point regarding the motion state; Substitute the value of the physical quantity of the vehicle regarding the motion state at the selected time point into the loss function to obtain the value of the loss function.

6. The speed planning method according to claim 3, characterized in that: at least one predicted movement route; and constructing a loss function of the vehicle according to the probability of the moving object entering the predicted movement route, comprising: Constructing a sub-loss function of the ego vehicle when the mobile object enters each predicted motion path; The sum of the products of the probability of the mobile object entering each predicted motion path and the corresponding sub-loss function is used as the loss function of the vehicle.

7. The speed planning method according to claim 6, characterized in that: The loss function also includes a traffic efficiency loss function; when constructing the sub-loss function of the vehicle when the mobile object enters each predicted motion path, it includes: When the mobile object enters each predicted motion route, the sum of the traffic efficiency loss function, the comfort loss function and the safety loss function of the ego vehicle is used as the sub-loss function of the ego vehicle.

8. The speed planning method according to claim 7, characterized in that: The traffic efficiency loss function is constructed as follows: According to the real-time status data, obtaining the real-time speed and real-time position of the vehicle; Determining a first difference between the real-time speed and a preset expected speed and a second difference between the real-time position and a preset expected position; Constructing the traffic efficiency loss function according to the first difference and the second difference and a preset speed weight coefficient and a position weight coefficient; The comfort loss function is constructed as follows: According to the real-time status data, obtaining the real-time acceleration and real-time jerk of the vehicle; Constructing the comfort loss function according to the real-time acceleration and the real-time jerk and the preset acceleration weight coefficient and jerk weight coefficient; The security loss function is constructed as follows: Acquire the real-time position of the vehicle and the real-time distance between the vehicle and the moving object according to the real-time perception data and the real-time status data; determining a third difference between the real-time distance and the safety distance; The safety loss function is constructed based on the real-time position, the third difference and a preset safety weight coefficient.

9. The method for speed planning according to any one of claims 1 to 8, characterized in that: If it is predicted that the moving object has the possibility of interfering with the ego vehicle, then according to the real-time perception data and the real-time status data, planning a real-time speed of the ego vehicle so as to keep at least a safe distance from the moving object includes: If it is predicted that the moving object has the possibility of interfering with the vehicle, judging whether the vehicle is currently traveling on a road section where overtaking is not allowed based on the real-time status data; If yes, planning a real-time speed of the ego vehicle to maintain at least a safe distance behind the moving object according to the real-time perception data and the real-time status data; If not, the real-time vehicle speed of the vehicle is planned to maintain at least a safe distance in front of the moving object according to the real-time perception data and the real-time status data.

10. A speed planning device, applied to a vehicle, characterized in that: include: An acquisition module, used to acquire real-time perception data of a vehicle's perception device on a moving object around the vehicle and real-time status data of the vehicle; wherein the moving object is an object moving relative to the ground; A prediction module, used for predicting the possibility that the moving object enters the front of the lane where the ego vehicle is located and interferes with the ego vehicle according to the real-time perception data; The planning module is used to plan the real-time speed of the ego vehicle to maintain at least a safe distance from the moving object based on the real-time perception data and the real-time status data if it is predicted that the moving object has the possibility of interfering with the ego vehicle.

11. An electronic device, characterized in that: The method comprises one or more processors, wherein the processors are used to implement the speed planning method according to any one of claims 1 to 9.

12. A vehicle, characterized in that: The method comprises one or more processors, wherein the processors are used to implement the speed planning method according to any one of claims 1 to 9.

13. A computer-readable storage medium, characterized in that: A computer program is stored thereon, and when the computer program is executed by a processor, the speed planning method according to any one of claims 1 to 9 is implemented.

Images