A speed planning method based on iterative optimization, a computer device, a readable storage medium and a motor vehicle

Abstract

The application discloses a speed planning method based on iterative optimization, a computer device, a readable storage medium and a motor vehicle, relates to the technical field of automatic driving, and is used for planning the speed of a vehicle when the vehicle runs in an automatic driving mode, and comprises the following steps: acquiring a planned path of the vehicle and a predicted trajectory of an obstacle around the vehicle to generate an obstacle boundary; generating a speed optimization function according to the obstacle boundary; iteratively solving the speed optimization function; and generating a speed curve according to the iterative solution. The application can ensure the operation efficiency while considering various constraint limitations.

Description

Technical Field

[0001] This invention relates to the field of autonomous driving technology, and specifically to an iterative optimization-based speed planning method, a computer device, a readable storage medium, and a motor vehicle. Background Technology

[0002] In the field of autonomous driving, the planning problem can often be broken down into two parts: lateral path planning and longitudinal velocity planning. Lateral path planning generates feasible paths, while longitudinal velocity planning dynamically avoids obstacles based on the lateral path planning. Velocity planning requires consideration of numerous constraints, including safety constraints from surrounding obstacles, vehicle dynamics constraints, and traffic rule constraints. It also needs to consider limitations such as driving comfort, user-set speed, and following distance. Therefore, velocity planning in autonomous driving needs to achieve the objective as much as possible while accommodating all constraints. Furthermore, nonlinear constraints are easily introduced into the various constraints of velocity planning in autonomous driving. Since nonlinear optimization solutions are prone to local minima and convergence difficulties, the computation time and complexity increase significantly, affecting the real-time performance of velocity planning. Summary of the Invention

[0003] This invention aims to address one of the technical problems in related technologies to a certain extent. To this end, this invention provides a speed planning method that ensures computational efficiency while taking into account various constraints.

[0004] To achieve the above objectives, the present invention adopts the following technical solution:

[0005] A speed planning method based on iterative optimization is used to plan the speed of a vehicle when it is operating in autonomous driving mode, including the following steps:

[0006] Obtain the planned path of the vehicle and the predicted trajectory of obstacles around the vehicle to generate obstacle boundaries;

[0007] Generate a velocity optimization function based on obstacle boundaries;

[0008] The speed optimization function is solved iteratively.

[0009] The velocity curve is generated based on the iterative solution.

[0010] Optionally, obtaining the planned path of the vehicle and the predicted trajectory of obstacles around the vehicle to generate obstacle boundaries includes the following steps:

[0011] Obtain the vehicle's planned path and the predicted trajectory of obstacles around the vehicle;

[0012] Traverse the predicted trajectory, determine whether the points in the predicted trajectory intersect with the planned path, and project the intersecting points onto the path-time image to form an obstacle quadrilateral.

[0013] The obstacle boundary is generated using the following formula:

[0014] S u (t i )=min{s max ,s vj (t i ),s vj+1 (t i ),s vj+2 (t i )...}-s buffer

[0015] S u (t i+1 )=min{s max ,s vj (t i+1 ),s vj+1 (t i+1 ),s vj+2 (t i+1 )...}-s buffer

[0016] Among them, S u (t i ) for t i Obstacle boundary at time s max s represents the maximum value along the planned path. vj (t i (v) represents the quadrilateral obstacle. j In t i The lower boundary of time, s buffer This is for redundancy;

[0017] The redundancy is calculated according to the following formula:

[0018] S buffer =S front_edge_to_center +S safe_distance ...

[0020] Among them, S front_edge_to_center S is the distance from the front of the car to the center of the rear axle. safe_distance This refers to the safe distance between the vehicle and the obstacle.

[0021] Optionally, generating a velocity optimization function based on obstacle boundaries includes the following steps:

[0022] Convert the velocity curve into an optimization variable vector;

[0023] Determine the speed optimization objective and obtain the optimization cost function;

[0024] Determine the cluster of function constraints based on the constraints.

[0025] Optionally, converting the velocity curve into an optimization variable vector includes the following steps:

[0026] Sampling is performed at time t with a sampling interval of Δt, dividing the time domain into n segments. The velocity curve s(t) is then divided into n velocity curves of length Δt. The resulting velocity curves are:

[0027]

[0028] Where k = 1…n, s k 'for s k The first-order differential term, s k "for s" k The second-order differential term, s k "'for s k The third-order differential term;

[0029] Each velocity curve segment is constrained according to the following formula:

[0030]

[0031] The n velocity curves are converted into an optimization variable vector according to the following formula:

[0032] x=(s1,s2,...,s n ,s1',s2',...,s n ',s1”,s2”,...,s n ”) T

[0033] Where x is the optimization variable vector.

[0034] Optionally, determining the speed optimization objective and obtaining the optimization cost function includes the following steps:

[0035] The comfort optimization target is determined according to the following formula:

[0036]

[0037] Where f1(x) is the comfort optimization objective, ω s "ω represents the weight of the amplitude of the brake and accelerator pedals." s "' is the weight of the push-back feeling;

[0038] The speed optimization target is determined based on the following formula:

[0039]

[0040] Where f2(x) is the velocity optimization objective, ω ref To set the weight of speed, S ref 'Set speed;'

[0041] The following formula is used to determine the target for optimizing following distance:

[0042]

[0043] Where f3(x) is the following distance optimization objective, D follow For following distance, ω follow Weighting for following distance;

[0044] The optimization cost function is determined according to the following formula:

[0045] f(x) = f1(x) + f2(x) + f3(x)

[0046] Here, f(x) is the optimization cost function.

[0047] Optionally, determining the family of function constraints based on the restrictions includes the following steps:

[0048] Obstacle constraints are determined using the following formula:

[0049] S k -S u (k)≤0 k=0,1,...,n-1,n

[0050] The vehicle dynamics constraints are determined according to the following formula:

[0051]

[0052] Among them, v max For the maximum speed of the vehicle, a max Let a be the maximum acceleration of the vehicle. min For the minimum acceleration of the vehicle, jerk max jerk is the maximum acceleration of the actuator. min This represents the minimum acceleration of the actuator.

[0053] Traffic rule constraints are determined using the following formula:

[0054]

[0055] Optionally, the velocity optimization function can be solved iteratively according to the following formula:

[0056]

[0057] in, This is the result obtained from the previous iteration of optimization. For approximate substitution of h(s) k A linear function of .

[0058] Optionally, the velocity curve can be generated according to the following formula:

[0059]

[0060] Where x is the optimization variable vector, v is the vehicle speed, a is the vehicle acceleration, and jerk is the acceleration of the vehicle's actuator.

[0061] The technical solution provided by this invention takes into account constraints such as safety constraints, vehicle power constraints, and traffic rule constraints in longitudinal speed planning, as well as limitations such as driving comfort, user-set speed, and following distance. Through the iteration and optimization of various constraints and limitations, it achieves a balance in all aspects without neglecting one aspect, which greatly promotes the development of autonomous driving technology.

[0062] Because the technical solution provided by this invention takes into account numerous constraints and limitations, it inevitably introduces nonlinear functions. Existing technologies either ignore nonlinear functions directly or linearize them through Taylor expansion. However, these methods are only applicable to continuous, analytically defined nonlinear functions, and are not suitable for complex autonomous driving scenarios. Furthermore, autonomous driving scenarios have strict real-time requirements for speed planning; each calculation must be completed within one cycle (a single cycle is often only 100ms). If the calculation exceeds the cycle, it will cause disorder in the speed planning and data interaction between upstream and downstream systems, affecting not only driving comfort but also potentially leading to traffic accidents. This invention addresses the issue of nonlinear functions by employing an iterative optimization method. In each iteration, the complex nonlinear problem is transformed into a linear problem, improving the applicability of nonlinear function processing. Simultaneously, without sacrificing accuracy, it ensures a linear solution to obtain a speed curve that satisfies all constraints, guaranteeing computational accuracy. Moreover, because this invention converts nonlinear functions into linear functions, it greatly simplifies computational complexity, shortens computation time, and ensures that each calculation is completed within one cycle.

[0063] Furthermore, the present invention also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the speed planning method described in any of the above-mentioned embodiments.

[0064] Meanwhile, a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the speed planning method described in any one of the above.

[0065] In addition, the present invention also provides a motor vehicle having an automatic driving function, wherein when the automatic driving function is in operation, the speed of the motor vehicle is planned by the speed planning method described in any one of the foregoing claims.

[0066] These features and advantages of the present invention will be disclosed in detail in the following specific embodiments and accompanying drawings. The preferred embodiments or means of the present invention will be shown in detail in conjunction with the accompanying drawings, but are not intended to limit the technical solutions of the present invention. In addition, each of these features, elements and components appearing in the following text and drawings is a plurality of, and different symbols or numbers are used for convenience of representation, but all represent parts with the same or similar construction or function. Attached Figure Description

[0067] The present invention will be further described below with reference to the accompanying drawings:

[0068] Figure 1 This is a schematic diagram of the planned path of the vehicle in an embodiment of the present invention;

[0069] Figure 2 This is a schematic diagram of the predicted trajectory of obstacles around the vehicle in an embodiment of the present invention;

[0070] Figure 3 This is a schematic diagram illustrating a collision between the vehicle and obstacles surrounding the vehicle in an embodiment of the present invention.

[0071] Figure 4 This is a schematic diagram of the region formed by expanding the line segment in an embodiment of the present invention;

[0072] Figure 5 This is a schematic projection of a collision between an obstacle around the vehicle and the vehicle in an embodiment of the present invention;

[0073] Figure 6 This is a schematic diagram of a real-world scenario for the planned path and predicted trajectory in an embodiment of the present invention;

[0074] Figure 7 This is a schematic diagram of the projection of obstacle boundary generation in an embodiment of the present invention;

[0075] Figure 8 This is a schematic diagram of a traffic constraint scenario in an embodiment of the present invention;

[0076] Figure 9 This is a projection diagram of traffic constraints in an embodiment of the present invention. Detailed Implementation

[0077] Embodiments of the present invention are described in detail below, examples of which are illustrated in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described are intended to explain the present invention and should not be construed as limiting the invention.

[0078] The terms "an embodiment," "example," or "trademark" used in this specification refer to a particular feature, structure, or characteristic described in connection with the embodiment itself that may be included in at least one embodiment disclosed in this patent. The phrase "in an embodiment" appearing in various places throughout the specification does not necessarily refer to the same embodiment.

[0079] Example:

[0080] This embodiment provides a speed planning method based on iterative optimization for speed planning of a vehicle during autonomous driving operations, including the following steps:

[0081] The planned path of the vehicle and the predicted trajectories of obstacles around the vehicle are obtained to generate obstacle boundaries. In this step, the planned path of the vehicle is obtained from the upstream planning module, such as... Figure 1 As shown, the planned path of the vehicle is to turn left at the T-junction, and the resulting planned path has N points P(s); the predicted trajectories of obstacles around the vehicle are obtained from the upstream prediction module, such as... Figure 2 As shown, v1, v2, v3, and v4 are obstacle vehicles at the T-junction, and the predicted trajectories obtained are N. p Point P p (t).

[0082] After obtaining this data, iterate through the predicted trajectory and determine whether any points in the predicted trajectory intersect with the planned path, such as... Figure 3 As shown, at prediction point Pp(t1), the planned path P(s) of the vehicle intersects with the predicted trajectory Pp(t) of the obstacle vehicle v1. Therefore, all predicted trajectories Pp(t) after t>t1 intersect with the planned path of the vehicle. Thus, the last prediction point is Pp(t2), intersecting with the vehicle at point P(s2). According to... Figure 3 Based on the judgment results, a two-dimensional path-time graph (st) is plotted with time (t) on the horizontal axis and path (s) on the vertical axis. The portion where the predicted trajectory of the obstacle intersects with the planned path of the vehicle is projected onto the path-time image. However, at this point, the image only shows line segments. Since both the vehicle and obstacle vehicles in reality have length and width, it is necessary to expand these line segments. The expansion of the region corresponding to each time step is as follows: Figure 4 As shown, Figure 4The diagram shows the determination of the shadow width at time t', which involves projecting the front and rear of the obstacle vehicle onto the path-time 2D st graph. This ultimately forms the obstacle quadrilateral.

[0083] like Figure 5 As shown, the black parallelogram area is... Figure 3 The projection result falling into the black area indicates that the vehicle will collide with the obstacle. The projection method is a common approach in this field for interaction between the vehicle's planning module and the obstacle prediction module, and will not be elaborated here.

[0084] like Figure 6 As shown, based on the aforementioned predictions, a schematic diagram of the actual scenario of the planned path and predicted trajectory is obtained, which is a schematic diagram of the actual scenario of the vehicle turning left at a T-junction. Following the aforementioned method, v2, v3, and v4 are projected onto the path-time image respectively, as follows... Figure 7 As shown, obstacle boundaries are generated according to the following formula: obstacle boundaries Figure 7 As shown by the dashed line:

[0085] S u (t i )=min{s max ,s vj (t i ),s vj+1 (t i ),s vj+2 (t i )...}-s buffer

[0086] S u (t i+1 )=min{s max ,s vj (t i+1 ),s vj+1 (t i+1 ),s vj+2 (t i+1 )...}-s buffer

[0087] Among them, S u (t i ) for t i Obstacle boundary at time s max s represents the maximum value along the planned path. vj (t i (v) represents the quadrilateral obstacle. j In t i The lower boundary of time, s buffer This is for redundancy. Redundancy is set to prevent the front of the vehicle from entering the shaded area. Redundancy is calculated using the following formula:

[0088] S buffer =S front_edge_to_center +S safe_distance ...

[0090] Among them, S front_edge_to_center S is the distance from the front of the car to the center of the rear axle. safe_distance This refers to the safe distance between the vehicle and the obstacle.

[0091] Specifically, in this embodiment, the obstacle boundary is:

[0092] s u (t i )=min{s max ,s v1 (t i ),s v2 (t i ),s v3 (t i ),s v4 (t i )}-s buffer

[0093] A speed optimization function is generated based on the obstacle boundary. This function is used to solve for the speed curve s(t). Its generation requires consideration of constraints such as safety constraints from surrounding obstacles, vehicle dynamics constraints, and traffic rule constraints. It also needs to consider limitations such as driving comfort, user-set speed, and following distance. In this embodiment, the optimization objective of the speed optimization function is to obtain the optimization variable with the minimum cost function under the combined constraints of the constraints and limitations. In speed optimization, the speed curve s(t) can be transformed into the optimization variable x. Constraints such as driving comfort, user-set speed, and following distance can be transformed into the optimization cost function f(x). The constraints such as safety constraints from surrounding obstacles, vehicle dynamics constraints, and traffic rule constraints can be represented by a family of constraint functions. Minimizing the cost function f(x) means that the speed curve s(t) satisfying all constraints can be solved. This step includes the following sub-steps:

[0094] Convert the velocity curve into an optimization variable vector:

[0095] Sampling is performed at time t with a sampling interval of Δt, dividing the time domain into n segments. The velocity curve s(t) is then divided into n segments of length Δt. Each segment of the velocity curve is represented by a third-order polynomial:

[0096]

[0097] Where k is a natural number, k = 1…n, s k 'for s kThe first-order differential term, s k "for s" k The second-order differential term, s k "'for s k The third-order differential term;

[0098] Since each velocity curve segment needs to satisfy the condition that the position, velocity, and acceleration at the connection point are equal, each velocity curve segment is constrained according to the following formula:

[0099]

[0100] Therefore, the variable that needs to be optimized for each segment of a third-order polynomial curve is s. k s k 'and s k "The n-segment third-order polynomial requires optimization of 3n variables. Based on this, the n-segment velocity curves are transformed into an optimization variable vector according to the following formula: "

[0101] x=(s1,s2,...,s n ,s1',s2',...,s n ',s1”,s2”,...,s n ”) T

[0102] Where x is the optimization variable vector.

[0103] Determine the speed optimization objective and obtain the optimization cost function. In this step, the speed optimization objective has three terms:

[0104] Comfort optimization goals, speed optimization goals, and following distance optimization goals.

[0105] The comfort optimization objective considers two aspects: the absolute values ​​of the second and third derivatives of the speed curve, and these two absolute values ​​should be minimized. Minimizing the absolute value of the second derivative of the speed curve reduces the amount of braking and accelerator pedal input required for straight-line driving; minimizing the absolute value of the third derivative of the speed curve reduces the driver's feeling of being pushed back in their seat, thus improving ride comfort. In the relevant formulas for speed optimization objectives, ω represents a weighting coefficient, which can be used to indicate the degree of importance placed on a particular objective term in the speed optimization objective.

[0106] The comfort optimization target is determined by the following formula:

[0107]

[0108] Where f1(x) is the comfort optimization objective, ω s "ω represents the weight of the amplitude of the brake and accelerator pedals." s "' is the weight of the push-back feeling;

[0109] In autonomous driving, the user sets a desired speed, and speed planning should aim to make the vehicle's speed as close as possible to the user's desired speed, minimizing the error between the two. Therefore, the speed optimization target is determined according to the following formula:

[0110]

[0111] Where f2(x) is the velocity optimization objective, ω ref To set the weight of speed, S ref 'Set speed;'

[0112] In autonomous driving scenarios, if there is an obstacle vehicle in front of the vehicle traveling at a certain speed, the autonomous driving function will require the vehicle to follow the obstacle vehicle at a certain distance. This following distance is set by the user; that is, the distance between the vehicle and the obstacle vehicle should be as close as possible to the set following distance. Therefore, the following distance optimization target is determined according to the following formula:

[0113]

[0114] Where f3(x) is the following distance optimization objective, D follow For following distance, ω follow The weight of following distance.

[0115] In this embodiment, the specific values ​​of each weight can be set by those skilled in the art according to the different positioning of various vehicle models, with different emphases on the push-back feeling, the magnitude of the brake and accelerator pedals, the user-set speed, and the user-set following distance. No limitation is made here.

[0116] Finally, considering the aforementioned objectives, the optimization cost function is determined according to the following formula:

[0117] f(x) = f1(x) + f2(x) + f3(x).

[0118] Here, f(x) is the optimization cost function.

[0119] Determine the function constraint family based on the constraints. In this step, obstacle constraints, vehicle dynamics constraints, and traffic rule constraints surrounding the vehicle must be considered.

[0120] Since obstacles around the vehicle have been filtered and projected onto the path-time image, and obstacle boundaries have been obtained, obstacle constraints should not exceed the obtained obstacle boundaries. Therefore, obstacle constraints are determined according to the following formula:

[0121] S k -S u (k)≤0, k=0,1,...,n-1,n

[0122] Vehicle dynamics constraints include maximum speed limits, actuator acceleration limits, and jerk limits. These constraints are determined using the following formula:

[0123]

[0124] Among them, v max For the maximum speed of the vehicle, a max Let a be the maximum acceleration of the vehicle. min For the minimum acceleration of the vehicle, jerk max jerk is the maximum acceleration of the actuator. min This represents the minimum acceleration of the actuator.

[0125] Regarding traffic rule constraints, such as Figure 8 As shown, the vehicle is preparing to go straight and enter the left-turn lane. The dashed line represents the planned path, s0 is the vehicle's starting position, s1 is the position where the vehicle just enters the curve, and s2 is the end of the planned path. Within the section from s0 to s1, the vehicle is affected by the road speed limit, with a maximum speed v. max For a speed of v1, after the interval from s1 to s2, the maximum speed v is limited by the road curvature. max Jump to v2, such as Figure 9 As shown, the relationship between speed and path is expressed using the function v. max =h(s k If we express this as an expression, then we obtain the following constraint formula for traffic rules:

[0126]

[0127] And the relational function v max =h(s k The specific form of the ( ) can be flexibly set by those skilled in the art according to the traffic rules of each region.

[0128] The technical solution provided in this embodiment takes into account constraints such as safety constraints, vehicle power constraints, and traffic rule constraints in longitudinal speed planning, as well as limitations such as driving comfort, user-set speed, and following distance. Through the iteration and optimization of various constraints and limitations, it achieves a balance in all aspects without neglecting one aspect, which greatly promotes the development of autonomous driving technology.

[0129] After establishing the constraints and optimization objectives, a solution is required. Since the technical solution provided in this embodiment considers numerous constraints and limitations, nonlinear functions are inevitably introduced. Existing technologies either ignore nonlinear functions directly or linearize them through Taylor expansion. However, these methods are only suitable for continuous, analytically expressible nonlinear functions, and are not applicable to complex autonomous driving scenarios. Furthermore, autonomous driving scenarios have strict real-time requirements for speed planning, meaning that a calculation must be completed within one cycle (a single cycle is often only 100ms). If the calculation exceeds the cycle, it will cause disorder in the speed planning and data interaction between upstream and downstream systems, affecting not only driving comfort but also potentially leading to traffic accidents. This embodiment provides an iterative solution method to address nonlinear problems, replacing nonlinear inequalities with linear inequalities in each iteration to ensure that each iteration is a linear optimization problem. The speed optimization function is iteratively solved according to the following formula:

[0130]

[0131] in, This is the result obtained from the previous iteration of optimization. For approximate substitution of h(s) k A linear function of ), based on this, in this iteration To determine constants to eliminate the effects of nonlinearity.

[0132] As can be seen, this embodiment adopts an iterative optimization method when dealing with nonlinear functions. In each iteration, the complex nonlinear problem is transformed into a linear problem, which improves the applicability of nonlinear function processing. At the same time, it ensures the linear solution of the problem without sacrificing accuracy, and obtains a velocity curve that meets all constraints, thus guaranteeing the accuracy of the calculation.

[0133] After linearization, the optimization problem in this embodiment is transformed into a standard quadratic optimization problem, which can be solved quickly using mature existing methods, such as the OSQP method. These are existing technologies and will not be elaborated upon here. After multiple iterations, the optimal variable x can be obtained. Because the nonlinear function is converted to a linear function before processing, the computational complexity is greatly simplified, the computation time is shortened, and each round of computation is ensured to be completed within one cycle.

[0134] Finally, the following formula is used to generate the speed curve based on the coefficients of the cubic polynomial obtained through iterative solution:

[0135]

[0136] Where x is the optimization variable vector, v is the vehicle speed, a is the vehicle acceleration, and jerk is the acceleration of the vehicle's actuator.

[0137] Meanwhile, this embodiment also provides a computer device, including a memory and a processor. The memory stores a computer program, and when the computer program is executed by the processor, it causes the processor to perform the steps of the speed planning method described above. The steps of the speed planning method here can be the steps in the memory analysis methods of the various embodiments described above.

[0138] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. Accordingly, the computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can implement the methods of any of the above embodiments. Any references to memory, storage, databases, or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), RAMbus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0139] Furthermore, this embodiment also provides a motor vehicle with an autonomous driving function. When the motor vehicle provided in this embodiment operates the autonomous driving function, it plans the speed of the vehicle using the aforementioned speed planning method.

[0140] The above are merely specific embodiments of the present invention, but the scope of protection of the present invention is not limited thereto. Those skilled in the art should understand that the present invention includes, but is not limited to, the contents described in the accompanying drawings and the specific embodiments above. Any modifications that do not depart from the functional and structural principles of the present invention will be included within the scope of the claims.

Claims

1. A speed planning method based on iterative optimization, used for speed planning of a vehicle during autonomous driving, characterized in that, The speed planning method includes the following steps: Obtain the planned path of the vehicle and the predicted trajectory of obstacles around the vehicle to generate obstacle boundaries; Generate a velocity optimization function based on obstacle boundaries; The speed optimization function is solved iteratively. The velocity curve is generated based on iterative solution; The process of generating a velocity optimization function based on obstacle boundaries includes the following steps: Convert the velocity curve into an optimization variable vector; Determine the speed optimization objective and obtain the optimization cost function; Determine the family of function constraints based on the restrictions; The process of converting the velocity curve into an optimization variable vector includes the following steps: Regarding time t Sampling is performed at intervals of 100°C. Δt Divide the time domain into n Segment, velocity curve s ( t ) divided into n Segment length is Δt The velocity curve, after segmentation, is: in, k =1… n , for The first-order differential term, for The second-order differential term, for The third-order differential term; Each velocity curve segment is constrained according to the following formula: According to the following formula n Transform the segment velocity curve into an optimization variable vector: in, x To optimize the variable vector.

2. The speed planning method according to claim 1, characterized in that, Obtaining the planned path of the vehicle and the predicted trajectories of obstacles around the vehicle to generate obstacle boundaries includes the following steps: Obtain the vehicle's planned path and the predicted trajectory of obstacles around the vehicle; Traverse the predicted trajectory, determine whether the points in the predicted trajectory intersect with the planned path, and project the intersecting points onto the path-time image to form an obstacle quadrilateral. The obstacle boundary is generated using the following formula: in, for Obstacle boundaries at any given moment The maximum value on the planned path. The obstacle is a quadrilateral exist The lower boundary of time, This is for redundancy; The redundancy is calculated according to the following formula: in, It is the distance from the front of the vehicle to the center of the rear axle. This refers to the safe distance between the vehicle and the obstacle.

3. The speed planning method according to claim 1, characterized in that, Determining the speed optimization objective and obtaining the optimization cost function involves the following steps: The comfort optimization target is determined according to the following formula: in, To optimize comfort, Weights for the amplitude of the brake and accelerator pedals. The weight of the push-back feeling; The speed optimization target is determined based on the following formula: in, To optimize the speed target, To set the weight of speed, To set the speed; The following formula is used to determine the target for optimizing following distance: in, To optimize following distance, To maintain following distance, Weighting for following distance; The optimization cost function is determined according to the following formula: in, To optimize the cost function.

4. The speed planning method according to claim 1, characterized in that, Determining the family of function constraints based on the restrictions includes the following steps: Obstacle constraints are determined using the following formula: The vehicle dynamics constraints are determined according to the following formula: in, The maximum speed of the vehicle. This is the maximum acceleration of the vehicle. The minimum acceleration of the vehicle. The maximum value of the actuator's acceleration. This represents the minimum acceleration of the actuator. Traffic rule constraints are determined using the following formula: 。 5. The speed planning method according to claim 4, characterized in that, The velocity optimization function is solved iteratively using the following formula: in, This is the result obtained from the previous iteration of optimization. For approximate substitution A linear function.

6. The speed planning method according to claim 5, characterized in that, The velocity curve is generated using the following formula: in, x To optimize the variable vector, v For the vehicle's speed, a For the vehicle's acceleration, jerk The acceleration of the actuator of the vehicle.

7. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the speed planning method according to any one of claims 1 to 6.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the speed planning method according to any one of claims 1 to 6.

9. A motor vehicle, characterized in that, The motor vehicle has an automatic driving function, and when the motor vehicle is operating the automatic driving function, it plans the speed of the vehicle using the speed planning method described in any one of claims 1 to 6.

Images