Parallel risk-based yielding strategy determination method and determination apparatus

Abstract

The application relates to the technical field of automatic driving, and provides a yielding strategy determination method and device based on parallel risks. The method obtains planning path information and environment information of a target mobile object, determines a risk mobile object based on the planning path information and the environment information, determines a defensive deceleration of the target mobile object relative to the risk mobile object, constructs a force model of the target mobile object based on at least the defensive deceleration and expected motion parameters of the target mobile object, determines an expected deceleration of the target mobile object based on the force model, and finally determines a yielding strategy of the target mobile object based on the expected deceleration. An effective yielding strategy of an automatic driving vehicle for parallel risks is proposed, and the efficiency of automatic driving vehicle decision-making is further improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of automatic driving, and in particular to a parallel risk-based yielding strategy determination method and device. BACKGROUND

[0002] During the operation of an autonomous vehicle, special attention should be paid to avoiding side-by-side driving with other vehicles. This is because long-term side-by-side driving will significantly reduce the lateral space of both vehicles, which may result in insufficient escape space between the two vehicles when encountering unexpected situations, increasing the risk of accidents. In addition, at a high vehicle speed, even a slight directional deviation may cause a dramatic change in the driving trajectory, thereby bringing higher scratching and collision risks to the side-by-side driving vehicles.

[0003] However, there is no good solution in the related art on how to develop a yielding strategy for an autonomous vehicle in view of parallel risk. SUMMARY

[0004] Therefore, the embodiments of the present application provide a parallel risk-based yielding strategy determination method and device to solve the problem of imperfect autonomous yielding strategy in the prior art.

[0005] In a first aspect, the embodiments of the present application provide a parallel risk-based yielding strategy determination method, comprising:

[0006] obtaining planning path information and environment information of a target mobile object;

[0007] determining a risk mobile object based on the planning path information and the environment information, the risk mobile object being a mobile object whose moving direction has an included angle less than a preset angle threshold with the moving direction of the target mobile object and whose predicted trajectory conflicts with the planning path of the target mobile object;

[0008] determining a defensive deceleration of the target mobile object relative to the risk mobile object;

[0009] constructing a force model of the target mobile object based at least on the defensive deceleration and expected motion parameters of the target mobile object;

[0010] determining an expected deceleration of the target mobile object based on the force model, and determining a yielding strategy of the target mobile object based on the expected deceleration.

[0011] In a second aspect, the embodiments of the present application provide a parallel risk-based yielding strategy determination device, comprising:

[0012] an obtaining module configured to obtain planning path information and environment information of a target mobile object;

[0013] The determining module is configured to determine a risk moving object based on the planned path information and the environment information, the risk moving object being a moving object similar in driving direction to the target moving object and having a conflict risk;

[0014] The determining module is further configured to determine a defensive deceleration of the target moving object relative to the risk moving object.

[0015] The constructing module is configured to construct a force model of the target moving object based on at least the defensive deceleration and expected motion parameters of the target moving object.

[0016] The determining module is further configured to determine an expected deceleration of the target moving object based on the force model, and determine a yielding strategy of the target moving object based on the expected deceleration.

[0017] Compared with the prior art, the embodiments of the present application have the beneficial effects that: the embodiments of the present application obtain planned path information and environment information of a target moving object, determine a risk moving object based on the planned path information and the environment information, determine a defensive deceleration of the target moving object relative to the risk moving object, construct a force model of the target moving object based on at least the defensive deceleration and expected motion parameters of the target moving object, determine an expected deceleration of the target moving object based on the force model, and finally determine a yielding strategy of the target moving object based on the expected deceleration, thus an effective yielding strategy of an autonomous vehicle for parallel risks is proposed, and the efficiency of decision-making of the autonomous vehicle is further improved. BRIEF DESCRIPTION OF DRAWINGS

[0018] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following will briefly introduce the drawings needed to be used in the embodiments or prior art description. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative effort on the basis of these drawings.

[0019] Figure 1 is a flowchart of a yielding strategy determination method based on parallel risks provided by the embodiments of the present application.

[0020] Figure 2 is a flowchart of a method for determining a risk moving object based on planned path information and environment information provided by the embodiments of the present application.

[0021] Figure 3 is a schematic diagram of another method for determining a risk moving object provided by the embodiments of the present application.

[0022] Figure 4 is a flowchart of a method for determining a defensive deceleration of a target moving object relative to a risk moving object provided by the embodiments of the present application.

[0023] Figure 5 is a schematic diagram of a motion trajectory of a target moving object and a risk moving object in a target coordinate system.

[0024] Figure 6 is a force analysis diagram of a target moving object.

[0025] Figure 7 is a force analysis diagram of a rearward moving object when the rear of the rearward moving object further includes two rearward moving objects.

[0026] Figure 8 is a force analysis diagram of a rearward moving object when the rear of the rearward moving object further includes multiple rearward moving objects.

[0027] Figure 9 is a schematic diagram of another parallel risk-based yielding strategy determination method.

[0028] Figure 10 is a schematic diagram of a parallel risk-based yielding strategy determination device.

[0029] Figure 11 is a schematic diagram of an electronic device. DETAILED DESCRIPTION

[0030] In the following description, specific details are set forth in order to provide a thorough understanding of embodiments of the application. However, persons having ordinary skill in the art will readily understand that embodiments of the application can be practiced without these specific details. In other instances, well-known structures, devices, circuits, and methods have not been described in detail in order to avoid obscuring the application.

[0031] A parallel risk-based yielding strategy determination method and device according to embodiments of the application will be described in detail below with reference to the accompanying drawings.

[0032] As mentioned above, during the operation of an autonomous vehicle, particular attention should be paid to avoiding side-by-side driving with other vehicles. In the related art, the state of side-by-side driving is usually broken by controlling the unmanned vehicle to appropriately decelerate or brake, forming a staggered structure, and then continuing to follow the vehicle. This processing method not only increases the lateral risk avoidance space, but also effectively reduces the risk of scratching caused by direction deviation, thereby improving driving safety. However, when controlling the unmanned vehicle to decelerate, how to determine the deceleration parameter, such as the deceleration, is a technical problem that needs to be solved.

[0033] In view of this, the embodiment of the present application provides a yielding strategy determination method based on parallel risk, which obtains the planning path information and the environment information of a target mobile object, determines a risk mobile object based on the planning path information and the environment information, determines the defensive deceleration of the target mobile object relative to the risk mobile object, constructs a force model of the target mobile object based on at least the defensive deceleration and the expected motion parameters of the target mobile object, determines the expected deceleration of the target mobile object based on the force model, and finally determines the yielding strategy of the target mobile object based on the expected deceleration, thereby providing an effective yielding strategy for the automatic driving vehicle in view of the parallel risk, and further improving the efficiency of the decision of the automatic driving vehicle.

[0034] Figure 1 FIG. 1 is a flowchart of a yielding strategy determination method based on parallel risk provided by the embodiment of the present application. As shown in the figure, the method comprises the following steps: Figure 1

[0035] In step S101, the planning path information and the environment information of a target mobile object are obtained.

[0036] In step S102, a risk mobile object is determined based on the planning path information and the environment information.

[0037] The risk mobile object is a mobile object whose moving direction has an included angle less than a preset angle threshold with the moving direction of the target mobile object, and whose predicted trajectory conflicts with the planning path of the target mobile object.

[0038] In step S103, the defensive deceleration of the target mobile object relative to the risk mobile object is determined.

[0039] In step S104, a force model of the target mobile object is constructed based on at least the defensive deceleration and the expected motion parameters of the target mobile object.

[0040] In step S105, the expected deceleration of the target mobile object is determined based on the force model, and the yielding strategy of the target mobile object is determined based on the expected deceleration.

[0041] In some embodiments of the present application, the method can be executed by a server or a terminal with certain computing capability to determine the yielding strategy of the target mobile object based on the parallel risk. The target mobile object can be an automatic driving vehicle, a robot or other mobile object, which is not limited here.

[0042] ​In some embodiments of the present application, the planning path information of the target mobile object and the environment information can be acquired first. The planning path information can be acquired from the path planning module of the target mobile object, and the environment information can be collected by the target mobile object itself, or collected by the server from the target mobile object, other traffic participants, and roadside devices, and then distributed to the target mobile object, or collected by the roadside device itself or distributed to the target mobile object from the server or other traffic participants, which is not limited here.

[0043] In some embodiments, the risk mobile object can be determined based on the planning path information and the environment information. The risk mobile object is a mobile object whose moving direction has an included angle with the moving direction of the target mobile object less than a preset angle threshold, and whose predicted trajectory conflicts with the planning path of the target mobile object. That is, the risk mobile object can be regarded as a "side-by-side" obstacle approximately parallel to the ego vehicle.

[0044] In some embodiments, after the risk mobile object is determined, the defensive deceleration of the target mobile object relative to the risk mobile object can be determined to avoid the conflict risk between the two. Further, a force model of the target mobile object can be constructed based on the determined defensive deceleration and the expected motion parameters of the target mobile object, the expected deceleration of the target mobile object is determined according to the force model, and the yielding strategy of the target mobile object is determined based on the expected deceleration. For example, the target mobile object is controlled to travel at the determined expected deceleration.

[0045] According to the technical solutions provided by the embodiments of the present application, by acquiring the planning path information and the environment information of the target mobile object, determining the risk mobile object based on the planning path information and the environment information, determining the defensive deceleration of the target mobile object relative to the risk mobile object, constructing the force model of the target mobile object based on at least the defensive deceleration and the expected motion parameters of the target mobile object, determining the expected deceleration of the target mobile object based on the force model, and finally determining the yielding strategy of the target mobile object based on the expected deceleration, an effective yielding strategy for the parallel risk of the autonomous vehicle is proposed, and the efficiency of the decision-making of the autonomous vehicle is further improved.

[0046] In some embodiments of the present application, the environment information at least includes a side-by-side moving object of the target moving object, the side-by-side moving object being a moving object with an included angle between a moving direction and a moving direction of the target moving object being less than a preset angle threshold. It can be understood that the side-by-side moving object to be considered can be a moving object with a distance to the target moving object being less than a preset threshold, for example, a moving object with a tail of an outer contour line being less than a first preset threshold from a head of an outer contour line of the target moving object, and a moving object with a head of the outer contour line being less than a second preset threshold from a tail of the outer contour line of the target moving object, where the distance can be a distance absolute value, and the first preset threshold and the second preset threshold can be set according to actual needs, which are not limited here.

[0047] Figure 2 is a flowchart of a method for determining a risk moving object based on planning path information and environment information provided by an embodiment of the present application. As shown in Figure 2 , the method comprises the following steps:

[0048] In step S201, the target moving object is represented by n decomposition circles with the same radius.

[0049] where n is a positive integer.

[0050] In step S202, starting from the center of each decomposition circle, n line segments are drawn in the normal direction of the moving direction of the target moving object, and the length of each line segment is a preset distance threshold.

[0051] In step S203, a moving object intersecting at least one of the n line segments with the outer contour line is determined as a risk moving object.

[0052] In some embodiments of the present application, the moving objects can be divided into three categories: forward moving objects on the moving path, backward moving objects on the moving path, and surrounding interactive moving objects. We define the risk moving object as a moving object similar to the moving direction of the target moving object and will cause a conflict in the future. For the obstacles that have conflict information with the moving path of the target moving object in the current state, the existing method can be well handled, and for the "side-by-side" obstacles approximately parallel to the moving direction of the target moving object, the longitudinal distance needs to be appropriately reduced to ensure driving safety.

[0053] The target moving object is represented by n decomposition circles with the same radius, and then starting from the center of each decomposition circle, n line segments are drawn in the normal direction of the moving direction of the target moving object, and the length of each line segment is a preset distance threshold. Finally, a moving object intersecting at least one of the n line segments with the outer contour line can be determined as a risk moving object.

[0054] Figure 3is a schematic diagram of another method for determining a risk mobile object provided by an embodiment of the present application. As shown in Figure 3 , a target mobile object is represented by three decomposition circles with the same radius r, and the centers of the three decomposition circles are o1, o2 and o3 respectively. The length of the target mobile object is l, the width is w, and the tangent direction of the driving direction is , the normal direction is , and the decomposition circle radius is , the distance between the centers of adjacent decomposition circles is

[0055] Since the risk of the inner side-by-side mobile object is usually small, the present embodiment only considers the driving track outside the target mobile object when determining the risk mobile object. As shown in Figure 3 , the outside of the target mobile object includes side-by-side mobile objects A, B and C. Three line segments are drawn from the centers o1, o2 and o3 to the outside of the driving track of the target mobile object along the normal direction, and the lengths of the three line segments are the preset distance threshold, then the line segments drawn from the centers o2 and o3 can be obtained, and the line segments intersect with the outer contour of the mobile object B at points b2 and b1 respectively, at this time it can be determined that the mobile object B is a risk mobile object.

[0056] Figure 4 is a flowchart of a method for determining the defensive deceleration of a target mobile object relative to a risk mobile object provided by an embodiment of the present application. As shown in Figure 4 , the method comprises the following steps:

[0057] In step S401, a target coordinate system is determined.

[0058] The horizontal coordinate axis of the target coordinate system is the tangent of the tail contour line of the risk mobile object after driving for a yielding time τ, and the vertical coordinate axis of the target coordinate system is the tangent of the current tail contour line of the risk mobile object.

[0059] In step S402, the formula is solved to obtain the defensive deceleration a s .

[0060] In some embodiments, a target coordinate system can be determined first, the horizontal coordinate axis of the target coordinate system is the tangent of the tail contour line of the risk mobile object after driving for a yielding time τ, and the vertical coordinate axis of the target coordinate system is the tangent of the current tail contour line of the risk mobile object. The value of the yielding time τ is determined by a path planning module, which will not be described here.

[0061] Figure 5 is a schematic diagram of the motion track of a target mobile object and a risk mobile object in a target coordinate system provided by an embodiment of the present application. As shown in Figure 5As shown, the target mobile object moves from the position of the solid line frame 1 to the position of the dashed line frame 1 after time τ, and the risk mobile object moves from the position of the solid line frame 2 to the position of the dashed line frame 2 after time τ. The horizontal coordinate axis of the target coordinate system is the tangent of the tail contour line of the dashed line frame 2, and the vertical coordinate axis is the tangent of the tail contour line of the solid line frame 2.

[0062] Taking the origin of the target coordinate system as the center, the motion trajectory of the target mobile object is a first circular arc, and the motion trajectory of the risk mobile object is a second circular arc. Let the central angle difference between the current risk mobile object and the target mobile object be α0, the central angle corresponding to the trajectory of the target mobile object in the yielding time τ be dα ego , and the central angle corresponding to the trajectory of the risk mobile object in the yielding time τ be dα agent , then α0+(dα agent -dα ego )≥0 needs to be satisfied. Further, the distance traveled by the target mobile object in the yielding time τ is , and the distance traveled by the risk mobile object in the yielding time τ is

[0063] That is, the following needs to be satisfied where Δs is the distance from the tail of the current risk mobile object to the head of the target mobile object in the target coordinate system, that is, in the target coordinate system, a first tangent is drawn from the origin to the tail contour line of the current risk mobile object, a second tangent is drawn from the origin to the tail contour line of the target mobile object, and the distance between the motion trajectory of the risk mobile object and the two tangents is denoted as Δs.v agent is the current speed of the risk mobile object, v ego is the current speed of the target mobile object, k is the curvature of the current travel path of the target mobile object, and d is the lateral distance of the target mobile object.

[0064] Further, the yielding time τ = β d -t0, where β is the yielding intensity, and different vehicle types have different yielding intensities, for example, large vehicles have a default value of 1.3, and small vehicles have a default value of 1.5. The default value of t0 may be, for example, 1.0.

[0065] Using this method, the defensive deceleration a s of the target mobile object relative to the risk mobile object can be determined. Further, a force analysis can be performed on the target mobile object, and the deceleration of the target mobile object can be further optimized based on the force analysis to obtain the desired deceleration.

[0066] In some embodiments of the present application, a force analysis can be performed on a target mobile object using a social force model. The social force model is a computational model that uses the concept of social forces to describe the interactions between traffic participants. The social force model takes into account the influence of the environment and other traffic participants and describes how the expected motion changes with these influences.

[0067] For a target traffic participant, according to the social force model, the total force it experiences is the sum of the attractive force of the expected motion, the repulsive force of other traffic participants, and the repulsive force of the road environment. For example, if the attractive force of the expected motion of the target traffic participant i is denoted as F i attr then where m i is the mass of the target traffic participant i, is the expected speed of i, is the expected motion direction of i, v i is the current speed of i, τ i is the relaxation time, i.e., the time interval needed to reach the expected speed and the expected motion direction.

[0068] For another example, if the repulsive force of other traffic participants experienced by the target traffic participant i is denoted as F i rep then where is the interaction repulsive force between the target traffic participant i and other traffic participant j, is the obstruction repulsive force between the target traffic participant i and static obstacle o, is the set of all moving obstacles, and O is the set of all static obstacles in the environment. Further, N is a positive integer, and i and j are positive integers less than or equal to N.

[0069] where the repulsive force between the target traffic participant i and obstacle k is denoted as F which is calculated by the formula where k ∈ P ∪ O, a k denotes the magnitude of the force, and b k denotes the range of the force, then r i,k = r i + r k , r i is the radius of the available rotationally invariant circular envelope of the target mobile object, r k is the radius of the available rotationally invariant circular envelope of obstacle k, and d ik is the Euclidean distance between the target mobile object and obstacle k. is the normalized vector pointing from obstacle k to the target mobile object. k is a positive integer less than or equal to N.

[0070] That is, the desired direction and speed of motion can not be maintained when other traffic participants or obstacles are present in the environment. In the social force model, the repulsive effect from these influences is described as an interaction force. This force prevents the traffic participant from moving in the intended direction, and it is modeled as the sum of forces introduced by other traffic participants or static obstacles.

[0071] In some embodiments, an anisotropy factor c ik characterizes the finiteness of the field of view of a human being, when

[0072] λ is the anisotropy strength. It is noted that this anisotropy factor is not needed in the autonomous driving scenario.

[0073] For example, if the road environment repulsive force F i env is used to represent the road environment repulsive force F wherein, c k is the repulsive force size, g(x) = x when x >= 0, otherwise g(x) = 0. P e is the set of all moving obstacles of the target moving object.

[0074] Thus, the resultant force F

[0075] Applying the social force model to the target moving object in the embodiments of the present application, the resultant force F

[0076] Figure 6 is the force analysis diagram of the target moving object provided by the embodiments of the present application. As Figure 6 shown, the target moving object has a force radius r e , the front moving object of the target moving object has a force radius r f , and the rear moving object of the target moving object has a force radius r r . The force F experienced by the target moving object includes a front attractive force F a rear repulsive force F

[0077] wherein, the rear repulsive force F The repulsive force generated by the backward moving object to the target moving object has the same direction as the attractive force of the forward moving object, so the resultant force on the target moving object can be determined as the sum of the attractive force of the forward moving object and the repulsive force of the backward moving object minus the difference of the repulsive forces of the surrounding moving objects. Further, the quotient of the resultant force on the target moving object and the mass of the target moving object is the expected deceleration.

[0078] In some embodiments of the present application, the attractive force of the target moving object can be determined by obtaining the expected motion parameters of the target moving object, the expected motion parameters including at least the expected speed v0, the expected acceleration a and the expected deceleration b of the target moving object. The attractive force of the target moving object can be calculated by the formula

[0079] wherein m is the mass of the target moving object, a is the forward following deceleration, b is the expected deceleration, and v is the current speed of the target moving object. ego f

[0080] wherein δ is an adjustable parameter, which can be set to an empirical value of 4 in an example, v is the current speed of the target moving object, Δv is the speed difference between the target moving object and the forward moving object, the forward moving object is a moving object traveling in front of the target moving object and adjacent to the target moving object, and T is the expected following time interval, which can be set according to actual needs and is not limited herein.

[0081] In some other embodiments of the present application, when determining the repulsive force of the surrounding moving objects on the target moving object, if only the repulsive force of the risk moving object on the target moving object is considered, the repulsive force of the target moving object can be calculated by the formula wherein m is the mass of the target moving object, and a is the defensive deceleration. ego s

[0082] In some further embodiments of the present application, the repulsive force of the backward moving object on the target moving object can be determined by calculating the forward repulsive force of the target moving object on the backward moving object wherein the backward moving object is a moving object traveling behind the target moving object and adjacent to the target moving object.

[0083] The forward repulsive force of the backward moving object can be calculated by force analysis on the backward moving object.

[0084] ​​​​​​​​​In some cases, there can be no mobile object within a preset distance behind the backward moving object. For example Figure 6 As shown in FIG. 1, there is no other mobile object behind the backward moving object, and the force acting on the backward moving object only includes the forward attractive force and the backward repulsive force generated by the target mobile object and the backward repulsive force generated by the target mobile object on the backward moving object and the forward repulsive force generated by the backward moving object on the target mobile object are equal in magnitude and opposite in direction, i.e.

[0085] At this time, the resultant force of the backward moving object can be first determined according to the mass of the backward moving object and the acceleration of the backward moving object. Then the forward attractive force of the backward moving object is calculated by using the above method of calculating the forward attractive force of the target mobile object. Finally, the forward repulsive force of the backward moving object can be obtained by subtracting the forward attractive force of the backward moving object from the resultant force of the backward moving object.

[0086] In other cases, there can be one or more backward moving objects within a preset distance behind the backward moving object, and different methods are needed to calculate the forward repulsive force of the backward moving object

[0087] If there is only one backward moving object within a preset distance behind the backward moving object, the forward repulsive force of the last mobile object is first obtained by using the above calculation method when there is no backward moving object within a preset distance behind the backward moving object.

[0088] Through analysis, it can be known that the resultant force acting on the backward moving object at this time includes the forward attractive force, the backward repulsive force generated by the last mobile object, and the forward repulsive force generated by the target mobile object.

[0089] Therefore, the resultant force acting on the last mobile object can be obtained by multiplying the mass of the last mobile object by the acceleration of the last mobile object, and then the forward repulsive force of the backward moving object can be obtained by subtracting the sum of the forward attractive force and the backward repulsive force from the resultant force

[0090] When there are multiple backward moving objects within a preset distance behind the backward moving object, the forward repulsive force of the current last backward moving object can be first calculated by using the force model that the resultant force is equal to the forward attractive force minus the forward repulsive force. The forward repulsive force of the last backward moving object is equal in magnitude and opposite in direction to the backward repulsive force of the second last backward moving object.

[0091] Then, the forward repulsive force of each mobile object is recursively calculated in sequence forwardly until the forward repulsive force of the backward mobile object is calculated by using the force model that the resultant force is equal to the forward attractive force plus the backward repulsive force minus the forward repulsive force.

[0092] That is, in response to determining that there are m backward mobile objects within the preset distance behind the backward mobile object, m is a positive integer, the forward repulsive force of the backward mobile object is The forward repulsive force of the mth backward mobile object is determined by using the method for calculating the forward repulsive force of the last backward mobile object, the reaction force of the forward repulsive force is the backward repulsive force of the (m-1)th backward mobile object; the resultant force of the (m-1)th backward mobile object is determined according to the mass of the (m-1)th backward mobile object and the acceleration of the (m-1)th backward mobile object; the forward attractive force of the (m-1)th backward mobile object is determined by using the method for calculating the target mobile object; the sum of the forward attractive force of the (m-1)th backward mobile object and the backward repulsive force of the (m-1)th backward mobile object is the forward component; the difference between the forward component and the resultant force of the (m-1)th backward mobile object is the forward repulsive force of the (m-1)th backward mobile object; the forward repulsive force of each backward mobile object is iteratively calculated in sequence from backward to forward by using the method for calculating the forward repulsive force of the (m-1)th backward mobile object, and the forward repulsive force of the backward mobile object is obtained.

[0093] Figure 7 is a force analysis schematic diagram provided by the embodiment of the present application when the backward mobile object further includes two backward mobile objects behind it. As Figure 7 shown. The backward mobile object 0 further includes two backward mobile objects, mobile object 1 and mobile object 2, the mobile object 1 and the mobile object 2 have no mobile object behind them, and the mobile object 1 and the mobile object 2 are not located on a straight line with the backward mobile object 0.

[0094] When the force of the backward mobile object 0 is analyzed, the mobile object 1 and the mobile object 2 can be analyzed first to determine the backward repulsive force of the mobile object 1 and the mobile object 2 on the backward mobile object 0.

[0095] For the mobile object 1, the resultant force thereof is wherein, is the mass of the mobile object 1, is the acceleration (or deceleration) of the mobile object 1. The mobile object 1 is subjected to the forward attractive force of the backward mobile object 0, and the backward repulsive force of the mobile object 2 The forward repulsive force of the mobile object 1 can be calculated by using the formula .

[0096] For the mobile object 2, the resultant force thereof is in, For the mass of the moved object 2, This is the acceleration (or deceleration) of moving object 2. Moving object 2 is subjected to a forward gravitational force. The forward repulsive force generated by the backward-moving object 0 and the backward repulsive force generated by the moving object 1 on it. in, Formulas can be used Calculations show that and It is a pair of forces that are equal in magnitude and opposite in direction.

[0097] By solving the force models of moving object 1 and moving object 2 separately, the following can be calculated: 5 and The value of .

[0098] For object 0 that moves backward, the net force acting on it is: ,in, The mass of object 0 is the mass that is moved backward. This represents the acceleration (or deceleration) of the backward-moving object 0. The backward-moving object 0 is subject to a forward gravitational force. The backward repulsive force generated by moving object 1 The backward repulsive force generated by the moving object 2 on it And the forward repulsive force exerted on it by the moving target object. in, Formulas can be used Calculations show that and They are a pair of forces of equal magnitude and opposite direction. and These are a pair of forces of equal magnitude and opposite direction. By solving the force model of the backward-moving object 0, its forward repulsive force can be calculated.

[0099] Figure 8 This is a schematic diagram of force analysis when there are multiple backward-moving objects behind the backward-moving object, as provided in the embodiments of this application. Figure 8 As shown. The backward-moving object 0 also includes m backward-moving objects, moving objects 1 to m, with force radii of r0, r1, ... r1 respectively. m In this context, the moving object m is the last moving object, and the backward moving object 0 is on a straight line with the other m backward moving objects.

[0100] At this point, we can first perform a force analysis on the moving object m. The net force acting on the moving object m is... in, is the mass of the mobile object m, is the acceleration (or deceleration) of the mobile object m. The mobile object m is subjected to a forward attractive force and a forward repulsive force generated by the mobile object m-1 wherein, can be calculated by the formula The force model of the mobile object m can be solved, and

[0101] Next, the force analysis of the mobile object m-1 is performed. The mobile object m-1 is subjected to a resultant force wherein, is the mass of the mobile object m-1, is the acceleration (or deceleration) of the mobile object m-1. The mobile object m-1 is subjected to a forward attractive force a backward repulsive force generated by the mobile object m and a forward repulsive force generated by the mobile object m-2 wherein, can be calculated by the formula , and are a pair of forces with equal magnitude and opposite directions. The force model of the mobile object m-1 can be solved, and

[0102] The mobile object m-2, m-3, …, is iteratively calculated in sequence, and the forward repulsive force generated by the previous mobile object on the mobile object can be calculated. Finally, the forward repulsive force generated by the target mobile object on the 0th backward mobile object can be calculated

[0103] That is, when the iteration reaches the mobile object 1, the mobile object 1 is subjected to a resultant force wherein, is the mass of the mobile object 1, is the acceleration (or deceleration) of the mobile object 1. The mobile object 1 is subjected to a forward attractive force a backward repulsive force generated by the mobile object 2 and a forward repulsive force generated by the mobile object 0 wherein, can be calculated by the formula , and are a pair of forces with equal magnitude and opposite directions. The force model of the mobile object 1 can be solved, and

[0104] For the backward mobile object 0, the resultant force subjected to is wherein, is the mass of the moving object 0, is the acceleration (or deceleration) of the moving object 0. The moving object 0 is subjected to a forward attractive force is the backward repulsive force generated by the moving object 1 on the moving object 0 and a forward repulsive force generated by the target moving object on the moving object 0 wherein, can be calculated by the formula , and are a pair of forces with equal magnitude and opposite direction. By solving the force model of the moving object 0, the

[0105] The embodiments of the present application provide a method for calculating the forward repulsive force generated by the target moving object on the backward moving object in various situations, which can select the appropriate method to calculate the optimal forward repulsive force according to the current road traffic situation. Further, the forward repulsive force can be used to determine the backward repulsive force of the target moving object, and the resultant force can be obtained by combining the forward attractive force and the surrounding repulsive force of the target moving object. Finally, the resultant force divided by the mass of the target moving object can obtain the expected deceleration of the target moving object, and then the expected deceleration is used to determine the yielding strategy of the target moving object.

[0106] The technical scheme of the embodiments of the present application proposes a simple and effective side-by-side driving risk identification and defensive deceleration calculation method. The method can identify the potential risks existing in side-by-side driving, and reduce the probability of potential accidents through timely deceleration measures, to provide more reliable safety protection for the automatic driving system

[0107] Further, the technical scheme provided by the embodiments of the present application converts the local yielding problem into a global optimization problem on the basis of considering the safety of surrounding vehicles, and realizes the effective application of the defensive deceleration strategy. By taking the yielding behavior into the optimization consideration of the overall traffic flow, the response ability of the automatic driving vehicle in complex road conditions is effectively improved, and the driving safety and traffic efficiency are further enhanced.

[0108] Figure 9 is a flowchart of another yielding strategy determination method based on parallel risk provided by the embodiments of the present application. As Figure 9As shown, if the target mobile object is an unmanned vehicle, the surrounding and ego vehicle information can be acquired first, which can include vehicle information in front of the ego vehicle, following vehicle information behind the ego vehicle, side-by-side vehicle information, and ego vehicle information, while acquiring ego vehicle planning path information. Next, identify vehicles at risk of side-by-side driving. Then calculate and implement a side-by-side misalignment deceleration strategy to limit the speed of side-by-side risk. Further analyze the force on the following vehicle behind the ego vehicle, infer the reaction force of the following vehicle on the ego vehicle, and perform forward recursion calculation to obtain backward expected maximum braking, speed limit repulsion and forward following attraction. Finally, perform force analysis to determine the longitudinal expected motion of the ego vehicle.

[0109] All the optional technical solutions described above can be combined to form optional embodiments of the present application, which will not be described one by one here.

[0110] The following is an embodiment of the device of the present application, which can be used to execute the method embodiments of the present application. For details not disclosed in the device embodiments of the present application, please refer to the method embodiments of the present application.

[0111] Figure 10 is a schematic diagram of a yielding strategy determination device based on parallel risk provided by an embodiment of the present application. As Figure 10 shown, the device includes:

[0112] The acquisition module 1001 is configured to acquire planning path information and environment information of a target mobile object.

[0113] The determination module 1002 is configured to determine a risk mobile object based on the planning path information and the environment information, the risk mobile object being a mobile object similar in driving direction to the target mobile object and having a conflict risk.

[0114] The determination module 1002 is further configured to determine a defensive deceleration of the target mobile object relative to the risk mobile object.

[0115] The construction module 1003 is configured to construct a force model of the target mobile object based on at least the defensive deceleration and expected motion parameters of the target mobile object.

[0116] The determination module 1002 is further configured to determine an expected deceleration of the target mobile object based on the force model, and determine a yielding strategy of the target mobile object based on the expected deceleration.

[0117] According to the technical scheme provided in the embodiments of the present application, the planning path information and the environment information of the target mobile object are acquired, the risk mobile object is determined based on the planning path information and the environment information, the defensive deceleration of the target mobile object relative to the risk mobile object is determined, the force model of the target mobile object is constructed based on at least the defensive deceleration and the expected motion parameter of the target mobile object, the expected deceleration of the target mobile object is determined based on the force model, and finally the yielding strategy of the target mobile object is determined based on the expected deceleration, thus an effective yielding strategy of the automatic driving vehicle for parallel risks is proposed, and the efficiency of the decision of the automatic driving vehicle is further improved.

[0118] In some embodiments, the environment information at least includes a side-by-side moving object of the target mobile object, and the side-by-side moving object is a mobile object with an included angle between a moving direction and a moving direction of the target mobile object less than a preset angle threshold; the risk mobile object is determined based on the planning path information and the environment information, including: representing the target mobile object by n decomposition circles with the same radius, n being a positive integer; drawing n line segments from the center of each decomposition circle along the normal direction of the moving direction of the target mobile object, each line segment having a preset distance threshold; and determining a mobile object with an intersection between an outer contour line and at least one of the n line segments as the risk mobile object.

[0119] In some embodiments, the defensive deceleration of the target mobile object relative to the risk mobile object is determined, including: determining a target coordinate system, a horizontal coordinate axis of the target coordinate system being a tangent of a tail outer contour line of the risk mobile object after a driving and yielding time τ, and a vertical coordinate axis of the target coordinate system being a tangent of a current tail outer contour line of the risk mobile object; and solving the formula to obtain the defensive deceleration a s ; wherein Δs is a distance from a tail of the current risk mobile object to a head of the target mobile object in the target coordinate system, v agent is a current speed of the risk mobile object, v ego is a current speed of the target mobile object, k is a curvature of a current driving path of the target mobile object, and d is a lateral distance of the target mobile object. wherein l is a length of the target mobile object, w is a width of the target mobile object, and r is a radius of the decomposition circle.

[0120]

[0121] In some implementations, the force model of the target moving object is as follows: the resultant force on the target moving object includes the forward attraction, the surrounding repulsive force, and the backward repulsive force of the target moving object; determining the desired deceleration of the target moving object based on the force model includes: determining that the resultant force on the target moving object is the sum of the forward attraction and the backward repulsive force minus the difference of the surrounding repulsive force; and determining that the quotient of the resultant force on the target moving object and the mass of the target moving object is the desired deceleration.

[0122] In some implementations, the forward gravitational force acting on the target moving object is determined as follows: the desired motion parameters of the target moving object are obtained, including at least the desired velocity v0, desired acceleration a, and desired deceleration b; and then the force is determined using the formula... The forward gravitational force of the target moving object is calculated. Where, m ego For the mass of the target moving object, a f To reduce speed when following the car in front, v is the current speed of the target moving object, and Δv is the speed difference between the target moving object and the forward moving object, which is the moving object that is moving in front of and adjacent to the target moving object.

[0123] In some implementations, the surrounding repulsive force on the target moving object is determined using the following method: [using a formula] Calculate the repulsive force around the target moving object. Where, m ego For the mass of the target moving object, a s For defensive deceleration.

[0124] In some implementations, the backward repulsive force on the target moving object is determined as follows: The forward repulsive force on the target moving object is calculated. Determine the forward repulsion force reaction force The backward repulsive force is the force acting on the target moving object; where the backward moving object is the moving object that is moving behind and adjacent to the target moving object.

[0125] In some implementations, in response to determining that there is no moving object within a preset distance behind the backward-moving object, the forward repulsive force of the backward-moving object... The following method is used to determine the following: the resultant force of the backward-moving object is determined based on its mass and acceleration; the forward attractive force of the backward-moving object is determined; the difference between the forward attractive force and the resultant force of the backward-moving object is determined as the forward repulsive force of the backward-moving object.

[0126] In some implementations, in response to determining that there are m backward-moving objects within a preset distance behind the backward-moving object, where m is a positive integer, the forward repulsive force of the backward-moving objects is... The following method is used to determine the forward repulsive force of the m-th backward-moving object; the reaction force of the forward repulsive force is determined as the backward repulsive force of the (m-1)-th backward-moving object; the resultant force of the (m-1)-th backward-moving object is determined based on its mass and acceleration; the forward attractive force of the (m-1)-th backward-moving object is determined; the sum of the forward attractive force and the backward repulsive force of the (m-1)-th backward-moving object is determined as the forward component; the difference between the forward component and the resultant force of the (m-1)-th backward-moving object is determined as the forward repulsive force of the (m-1)-th backward-moving object; using the above method for calculating the forward repulsive force of the (m-1)-th backward-moving object, the forward repulsive force of each backward-moving object is calculated iteratively from back to front to obtain the forward repulsive force of the backward-moving object.

[0127] It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0128] Figure 11 This is a schematic diagram of an electronic device provided in an embodiment of this application. Figure 11 As shown, the electronic device 11 of this embodiment includes: a processor 1101, a memory 1102, and a computer program 1103 stored in the memory 1102 and executable on the processor 1101. When the processor 1101 executes the computer program 1103, it implements the steps in the various method embodiments described above. Alternatively, when the processor 1101 executes the computer program 1103, it implements the functions of each module / unit in the various device embodiments described above.

[0129] Electronic device 11 may be a desktop computer, laptop, handheld computer, cloud server, or other electronic device. Electronic device 11 may include, but is not limited to, processor 1101 and memory 1102. Those skilled in the art will understand that... Figure 11 This is merely an example of electronic device 11 and does not constitute a limitation on electronic device 11. It may include more or fewer components than shown, or different components.

[0130] The processor 1101 can be a central processing unit (CPU), and can also be other general-purpose processors, a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, or the like.

[0131] The memory 1102 can be an internal storage unit of the electronic device 11, for example, a hard disk or a memory of the electronic device 11. The memory 1102 can also be an external storage device of the electronic device 11, for example, a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, or the like equipped on the electronic device 11. The memory 1102 can also include both the internal storage unit and the external storage device of the electronic device 11. The memory 1102 is used to store computer programs and other programs and data required by the electronic device.

[0132] Those skilled in the art can clearly understand that, for the convenience and brevity of description, only the division of the above functional units and modules is exemplified, and in actual application, the above functions can be completed by different functional units and modules according to needs, that is, the internal structure of the device is divided into different functional units or modules to complete all or part of the functions described above. Each functional unit and module in the embodiment can be integrated in one processing unit, or each unit can exist physically, or two or more units can be integrated in one unit, and the integrated unit can be realized in the form of hardware or in the form of a software functional unit.

[0133] The integrated modules / units, if implemented in the form of software functional units and sold or used as independent products, can be stored in a computer readable storage medium. Based on such understanding, all or part of the processes in the above-mentioned embodiment methods can also be completed by a computer program instructing related hardware, and the computer program can be stored in a computer readable storage medium. The computer program can be executed by a processor to implement the steps of the above-mentioned various method embodiments. The computer program can include computer program code, which can be in the form of source code, object code, executable files, or some intermediate forms. The computer readable medium can include any entity or device capable of carrying the computer program code, recording medium, U disk, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electric carrier signal, telecommunication signal, and software distribution medium, etc.

[0134] The above embodiments are only used to illustrate the technical solutions of the present application, rather than limit them; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacements for part of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application, and should be included in the protection scope of the present application.

Claims

1. A method for determining yielding strategies based on parallel risk, characterized in that, include: Obtain the planned path information and environmental information of the target moving object; Based on the planned path information and environmental information, risky moving objects are determined. The risky moving objects are moving objects whose moving direction is at an angle less than a preset angle threshold with the moving direction of the target moving object, and whose predicted trajectory conflicts with the planned path of the target moving object. Determine the defensive deceleration of the target moving object relative to the risky moving object; A force model of the target moving object is constructed based at least on the defensive deceleration and the expected motion parameters of the target moving object; The desired deceleration of the target moving object is determined based on the force model, and the yielding strategy of the target moving object is determined based on the desired deceleration.

2. The method according to claim 1, characterized in that, The environmental information includes at least the moving objects traveling side-by-side with the target moving object, wherein the moving objects traveling side-by-side are moving objects whose moving direction makes an angle less than a preset angle threshold with respect to the moving direction of the target moving object; The process of determining risky moving objects based on the planned path information and environmental information includes: The target moving object is represented by n decomposed circles with the same radius, where n is a positive integer; Starting from the center of each decomposed circle, draw n line segments along the normal direction of the moving target object's travel direction, with the length of each line segment being a preset distance threshold. The moving object whose outer contour line intersects with at least one of the n line segments is identified as the risk moving object.

3. The method according to claim 2, characterized in that, Determining the defensive deceleration of the target moving object relative to the risky moving object includes: A target coordinate system is defined, wherein the horizontal axis of the target coordinate system represents the time required for the moving object to make way. The tangent to the outer contour of the tail of the target coordinate system is the tangent to the outer contour of the current tail of the risk-moving object. Solving the formula The defensive deceleration a is obtained. s ; in, v is the distance from the tail of the currently moving risk object to the head of the moving target object in the target coordinate system. agent v is the current speed of the moving object at risk. ego Let k be the current speed of the target moving object, k be the curvature of the current travel path of the target moving object, and d be the lateral distance of the target moving object; Where l is the length of the target moving object, w is the width of the target moving object, and r is the radius of the decomposition circle. .

4. The method according to claim 1, characterized in that, The force model of the target moving object is as follows: the resultant force on the target moving object includes the forward attraction, the surrounding repulsive force, and the backward repulsive force of the target moving object. Determining the desired deceleration of the target moving object based on the force model includes: The resultant force acting on the target moving object is determined to be the sum of the forward attractive force and the backward repulsive force minus the difference of the surrounding repulsive force; The desired deceleration is determined by the ratio of the net force acting on the target moving object to the mass of the target moving object.

5. The method according to claim 4, characterized in that, The forward gravitational force acting on the target moving object is determined in the following manner: Obtain the desired motion parameters of the target moving object, wherein the desired motion parameters include at least the desired velocity v0, desired acceleration a, and desired deceleration b of the target moving object; Using formula The forward gravitational force of the target moving object was calculated. ; Where, m ego a is the mass of the target moving object. f To reduce speed when following the car in front, , , This is an adjustable parameter, where v is the current speed of the target moving object. The speed difference between the target moving object and the forward moving object, where the forward moving object is the moving object traveling in front of and adjacent to the target moving object, and T is the desired following distance.

6. The method according to claim 4, characterized in that, The surrounding repulsive force experienced by the target moving object is determined in the following manner: Using formula The repulsive force around the target moving object is calculated. , where m ego a is the mass of the target moving object. s This refers to the defensive deceleration.

7. The method according to claim 4, characterized in that, The backward repulsive force experienced by the target moving object is determined in the following manner: Calculate the forward repulsive force exerted by the target moving object on the backward moving object. ; Determine the forward repulsive force reaction force The backward repulsive force of the target moving object; The backward moving object is a moving object that travels behind and is adjacent to the target moving object.

8. The method according to claim 7, characterized in that, In response to determining that there is no moving object within a preset distance behind the backward-moving object, the forward repulsive force of the backward-moving object... The following method is used to determine: The resultant force of the backward-moving object is determined based on the mass and acceleration of the backward-moving object; Determine the forward gravitational force of the backward-moving object; The difference between the forward attractive force and the resultant force of the backward-moving object is determined as the forward repulsive force of the backward-moving object.

9. The method according to claim 7, characterized in that, In response to determining that there are m backward-moving objects within a preset distance behind the backward-moving object, where m is a positive integer, the forward repulsive force of the backward-moving objects... The following method is used to determine: Determine the forward repulsive force of the m-th backward-moving object, and determine the reaction force of the forward repulsive force as the backward repulsive force of the (m-1)-th backward-moving object; The resultant force of the (m-1)th backward-moving object is determined based on the mass and acceleration of the (m-1)th backward-moving object. Determine the forward gravitational force of the (m-1)th backward-moving object; The sum of the forward attraction force and the backward repulsion force of the (m-1)th backward-moving object is determined as the forward component; The difference between the forward component and the resultant force of the (m-1)th backward moving object is determined as the forward repulsive force of the (m-1)th backward moving object; The forward repulsion force of the backward-moving object is calculated by iteratively calculating the forward repulsion force of each backward-moving object from back to front, thus obtaining the forward repulsion force of the backward-moving object.

10. The method according to claim 9, characterized in that, Determining the forward repulsive force of the m-th backward-moving object includes: The resultant force of the m-th backward-moving object is determined based on the mass and acceleration of the m-th backward-moving object. Determine the forward gravitational force of the m-th backward-moving object; The difference between the forward attractive force and the resultant force of the m-th backward-moving object is determined as the forward repulsive force of the m-th backward-moving object.

11. The method according to any one of claims 8 to 10, determining the forward gravitational force of each backward-moving object, comprising: Obtain the desired motion parameters of each backward-moving object, and use the desired motion parameters to determine the forward following deceleration of each backward-moving object; Obtain the mass of each backward-moving object, and determine the product of the mass of the target backward-moving object and the forward following deceleration of the target backward-moving object as the forward gravitational force of the target backward-moving object; Wherein, the target backward-moving object is any backward-moving object.

12. A yielding strategy determination device based on parallel risk, characterized in that, include: The acquisition module is configured to acquire the planned path information and environmental information of the target moving object; The determination module is configured to determine risky moving objects based on the planned path information and environmental information. The risky moving objects are moving objects that have a similar driving direction to the target moving object and have a risk of conflict. The determining module is also configured to determine the defensive deceleration of the target moving object relative to the risky moving object; The building module is configured to build a force model of the target moving object based at least on the defensive deceleration and the desired motion parameters of the target moving object; The determining module is further configured to determine the expected deceleration of the target moving object based on the force model, and to determine the yielding strategy of the target moving object based on the expected deceleration.

Images