CN105857306A - Vehicle autonomous parking path programming method used for multiple parking scenes - Google Patents

Abstract

The invention provides a vehicle autonomous parking path programming method used for multiple parking scenes. The method is used for automatically parking a vehicle in a parking space through an autonomous parking system when the autonomous parking system detects the available parking space. The method includes the steps that target parking space information is detected, and a parking scene is determined; the initial state and target state of the to-be-parked vehicle are determined; a vehicle kinematics differential equation is established; state variables and control variables of the vehicle are segmented, equidistance sampling is performed on each segment according to certain time step, and to-be-optimized variables are obtained; an equality constraint, boundary constraints and inequality constraints of the to-be-optimized variables are formed; motion range constraints of the to-be-parked vehicle are formed according to the motion range limit in the parking process of the vehicle; an optimization objective is determined, and an objective function is established; and by means of a nonlinear programming solver, an optimal solution of a parking path is obtained. The vehicle autonomous parking path programming method is suitable for the multiple parking scenes, the design is reasonable, abundant information can be provided so as to control autonomous parking of the vehicle, and the security coefficient is high.

Description

Vehicle autonomous parking path planning method for multiple parking scenes

Technical Field

The invention relates to the technical field of vehicle autonomous parking, in particular to a vehicle autonomous parking path planning method for various parking scenes.

Background

In recent years, with the rapid increase of the quantity of automobiles kept in China, parking spaces in cities are increasingly tense and narrow. For a novice driver, parking is usually a difficult problem, particularly for the situation that the parking space is too narrow, the driver often has difficulty in well controlling the automobile to quickly and accurately park, and the probability of accidents caused by parking is greatly increased.

The autonomous parking system can help a driver to park accurately and safely, detect the size and the position of a parking space by using one or more sensors, plan a feasible parking path, and finally automatically control a steering system, a braking system and a power system of a vehicle to finish parking according to the planned path. In the autonomous parking system, path planning is one of the key technologies. Safe collision-free, path-feasible is the most basic and important requirement, and on this basis, a fast and comfortable parking path is also required by the autonomous parking system. In addition, if the parking path planning result can provide richer information to the execution system, the planned path can be tracked more conveniently.

The invention patent CN102975715A in China provides a method for planning parallel parking paths of an automobile in any posture, and the method traverses a spline curve fitted by a dot matrix connecting a start point and an end point of the automobile and then searches a path which accords with vehicle kinematics constraint and collision avoidance constraint. The method has the disadvantages that the planning path of the method assumes that the vehicle moves only in one direction, and the method does not meet the actual requirement of multiple adjustment in the parking process, so the planning success rate is not high.

A method for determining a vehicle path for an automatic parallel parking system is provided in the patent application No. 201210547981.2, which may provide parallel parking path planning for a single cycle steering maneuver or a two cycle steering maneuver. A method for forward parking a vehicle into a vertical parking space is provided in the patent application No. 201080064605.7. The disadvantage of the above method is that it is only suitable for planning a parking path of a parking space. In addition, the method does not provide information such as speed, acceleration and the like in the planning result, and is not beneficial to tracking the planned path.

The patent with the application number of 201510737989.9 provides a dynamic optimization framework of vehicle-environment integrated modeling based on a fully-simultaneous solution strategy, and effectively eliminates the influence of different parking space shapes on a trajectory planning strategy. Disadvantageously, the method does not restrict the moving range of the vehicle, and the planned track of the method can cause the vehicle to invade other lanes to prevent the vehicle on the other lanes from driving and even causing accidents. It is also disadvantageous that it does not detect whether a collision occurs between two discrete vehicle states, which may cause the vehicle to collide while traveling along the planned trajectory.

Disclosure of Invention

The invention aims to provide a vehicle autonomous parking path planning method for multiple parking scenes, which aims to solve the defects in the prior art.

The technical scheme of the invention is as follows:

a vehicle autonomous parking path planning method for a plurality of parking scenarios for an autonomous parking system to automatically park a vehicle in an available parking space upon detection of the parking space, comprising the steps of:

(1) detecting target parking space information and determining a parking scene;

(2) determining an initial state and a target state of a vehicle to be parked according to the target parking space information and the parking scene;

(3) establishing a vehicle kinematic differential equation based on an Ackerman model of the front-wheel steering four-wheel vehicle;

(4) segmenting the vehicle state variable and the control variable, and carrying out equidistant sampling on each segment according to a certain time step length to obtain a variable to be optimized;

(5) expressing the differential of each sampling point of the vehicle state variable as a function of each sampling point on a section where the sampling point is positioned by adopting a Lagrange interpolation method, and simultaneously connecting the function and a vehicle kinematic differential equation to convert the vehicle kinematic differential equation into an algebraic equation to form an equality constraint of the variable to be optimized;

(6) forming boundary constraints of variables to be optimized according to physical limitations of vehicle motion and safety requirements of parking;

(7) formulating collision avoidance requirements according to obstacles around a target parking space to form inequality constraints of variables to be optimized;

(8) forming a motion range constraint of the vehicle to be parked according to the motion range limit of the vehicle in the parking process;

(9) determining an optimization target and establishing an objective function;

(10) and obtaining the optimal solution of the parking path by adopting a nonlinear programming solver.

The vehicle autonomous parking path planning method for the multiple parking scenes further comprises the following steps of:

and (3) fitting the vehicle state variables in the optimal solution of the parking path by adopting a Lagrange interpolation method, sampling the fitted vehicle state variables by using a refined time step length to obtain a refined vehicle state sequence, detecting whether each vehicle state in the refined vehicle state sequence is collided, if so, increasing the number of segments, and repeating the steps (4) to (10) to plan the path again.

The vehicle autonomous parking path planning method for multiple parking scenes comprises the steps that in the step (1), the target parking space information comprises the direction, the position, the length and the width of a target parking space and the positions of obstacles around the target parking space; the parking scenarios include vertical parking, diagonal parking, and parallel parking.

In the method for planning the autonomous parking path of the vehicle for multiple parking scenes, in the step (3), the vehicle kinematic differential equation is as follows:

wherein x represents the abscissa of the vehicle rear axle center in a Cartesian coordinate system, and y represents the vehicle rear axle center in a Cartesian coordinate systemThe ordinate in the coordinate system, v represents the moving speed of the center of the rear axle of the vehicle, theta represents the angle between the orientation of the vehicle and the X-axis of the Cartesian coordinate system,representing the angle of rotation of the front wheels of the vehicle, a representing the acceleration of the centre of the rear axle of the vehicle, ω representing the angular velocity of the angle of rotation of the front wheels of the vehicle, L_mRepresenting the distance between the front axle and the rear axle of the vehicle;

in the step (4), the vehicle state variables are x, y, v, theta,The vehicle control variable is a and omega, the vehicle state variable and the control variable are segmented, the number of the segments is N, each segment comprises M sampling points at equal intervals, the time length of each segment is (M-1) h, and the variables x, y, v, theta and delta are measured by a computer,In the M sampling points of each segment after ω dispersion, the sampling points at both ends are shared with the adjacent segments, and each segment after variable a dispersion contains M independent sampling points, then the variable to be optimized is:

wherein h represents a time step; x is the number of_iI is 0, 1, …, (M-l) N represents the value of the ith sample point after the variable x is discretized; y is_iI is 0, 1, …, (M-1) N represents the value of the ith sample point after the dispersion of the variable y; v. of_iI is 0, 1, …, (M-1) N represents the value of the ith sample point after the dispersion of the variable v; theta_iWhere i is 0, 1, …, (M-1) N represents the value of the ith sample point after the variation theta is dispersed,representing variablesThe value of the ith sample point after dispersion; omega_iI is 0, 1, …, (M-1) N represents the value of the ith sample point after the variable ω is discretized; a is_pP is 0, 1, …, and MN-l indicates the value of the p-th sampling point after the dispersion of the variable a.

The vehicle autonomous parking path planning method for multiple parking scenes adopts a Lagrange interpolation method to carry out vehicle state variables x, y, v, theta,The differential at each sample point is expressed as a function of the M sample points over the segment in which the sample point lies:

$s_{n,m}^{\′}=\frac{1}{h}\underset{m=0}{\overset{M-1}{\mathrm{\Σ}}}{\mathrm{\γ}}_{n,m}*s_{n,m},n=0,1,...,N-1;m=0,1,\mathrm{...},M-1$

and combining the function and the vehicle kinematic differential equation to convert the vehicle kinematic differential equation into an algebraic equation to form an equation constraint of a variable to be optimized:

s′_n，m*h-f(t_n，m)*h＝0，n＝0，1，…，N-1；m＝0，1，…，M-1

wherein s represents a vehicle state variable x, y, v, θ,s_n，mRepresenting vehicle state variables at time t_n，mValue of (a), gamma_n，mRepresenting vehicle state variables at time t_n，mCoefficient of (d), s'_n，mRepresenting vehicle state variables at time t_n，mDerivative of (a), t_n，m＝h*[(M-1)*n+m]Indicating the time instant at the mth sampling point within the nth segment.

According to the vehicle autonomous parking path planning method for multiple parking scenes, the boundary constraint of the variable to be optimized is as follows:

wherein h is_maxMaximum limit value, x, representing a time step_lb、y_lb、v_lb、a_lb、θ_lb、ω_lbRespectively represent variables x, y, v, a, theta,Lower limit of ω, x_ub、y_ub、v_ub、a_ub、θ_ub、ω_ubRespectively represent variables x, y, v, a, theta,An upper limit value of ω;indicating an initial state of the vehicle to be parked,indicating a target state of the vehicle to be parked; x is the number of_i、y_i、v_i、θ_i、ω_iRespectively represent variables x, y, v, theta,Value of ith sample point after ω dispersion, a_pRepresenting the value of the p-th sample point after the dispersion of the variable a.

According to the vehicle autonomous parking path planning method for multiple parking scenes, inequality constraints of variables to be optimized are as follows:

$\left\{\begin{array}{cc}S(C_i,P_{j,k})>\mathrm{\α}*SA,& i=0,1,\mathrm{...},(M-1)N;j=1,2,\mathrm{...},J;k=1,2,3,4\\ S(P_j,C_{i,k})>\mathrm{\α}*{\mathrm{SP}}_j,& i=0,1,\mathrm{...},(M-1)N;j=1,2,\mathrm{...},J;k=1,2,3,4\end{array}\right.$

wherein, C_iRepresenting a quadrilateral into which the vehicle to be parked is abstracted when it is in the i-th state, P_jRepresenting a quadrilateral into which the jth obstacle is abstracted, J representing the number of obstacles, P_j，kRepresenting a quadrilateral P_jOf the kth corner point, C_i，kRepresents a quadrangle C_iThe kth corner of (2), S (C)_i，P_j，k) Represents P_j，kAnd a quadrangle C_iThe sum of the areas of the four triangles formed, SA, representing the quadrilateral C_iArea of (d), S (P)_j,C_i，k) Is represented by C_i，kAnd a quadrangle P_jSum of areas of four triangles formed, SP_jRepresenting a quadrilateral P_jα is a safety factor greater than 1.

According to the vehicle autonomous parking path planning method for multiple parking scenes, the motion range of a vehicle to be parked is constrained as follows:

$\left\{\begin{array}{cc}{x}_{lb}\≤x_{C_{i,k}}\≤x_{ub},& i=1,\mathrm{2...},(M-1)N-1;k=1,2,3,4\\ y_{lb}\≤y_{C_{i,k}}\≤y_{ub},& i=1,2,\mathrm{...},(M-1)N-1;k=1,2,3,4\end{array}\right.$

wherein,x-axis coordinates representing a kth corner point of a quadrangle into which the vehicle to be parked is abstracted when it is in the ith state,y-axis coordinates representing a kth angular point of a quadrangle abstracted when the vehicle to be parked is in the ith state; x is the number of_lb、x_ubRespectively representing the lower and upper limits of a vehicle state variable x, y_lb、y_ubRespectively representing the lower limit value and the upper limit value of the vehicle state variable y.

The vehicle autonomous parking path planning method for multiple parking scenes is characterized in that an objective function is as follows:

T_f＝N*(M-1)h

wherein, T_fRepresenting a path planning time of the vehicle to be parked;

the optimization target is the shortest time, minT_f。

According to the vehicle autonomous parking path planning method for multiple parking scenes, M is an integer and is more than or equal to 4 and less than or equal to 8.

The invention has the beneficial effects that:

according to the technical scheme, the method is suitable for path planning of various parking scenes, including vertical parking, oblique parking and parallel parking, can provide parking paths which accord with vehicle kinematic constraints and collision avoidance constraints, is safe and feasible in planning results, can provide control information such as speed and acceleration so as to facilitate tracking of the planned paths, is reasonable in design, can provide abundant information to control autonomous parking of the vehicle, and is high in safety factor.

Drawings

Fig. 1 is a block diagram of an autonomous parking system to which an embodiment of the present invention is applied;

FIG. 2 is a vehicle geometry schematic of an embodiment of the present invention;

FIG. 3 is a schematic diagram of a vertical parking path planning in accordance with an embodiment of the present invention;

FIG. 4 is a schematic diagram of the diagonal parking path planning according to the embodiment of the present invention;

fig. 5 is a schematic diagram of parallel parking path planning according to an embodiment of the present invention.

Detailed Description

The invention is further described with reference to the following figures and specific examples.

As shown in fig. 1, the autonomous parking system includes a sensing system 1, a controller 2, and a steering system 31, a braking system 32, and a power system 33 of the vehicle. The sensing system 1 includes one or more sensors, such as an ultrasonic-based sensor, a vision-based sensor, or a laser-based sensor, which can detect information of surrounding obstacles and detect parking space information, set a vehicle target state, and then transmit the above information to the controller 2. The controller 2 receives the obstacle information, the target parking space information and the target state information sent by the sensing system 1, then models and solves the parking problem according to the method of the invention, and finally executes the planned path. Steering system 31, braking system 32, and powertrain 33 may receive and execute control commands from controller 2 and send feedback information to controller 2. For example, steering system 31 may receive a steering wheel angle command or a vehicle front wheel steering angle command and implement a corresponding steering wheel angle or vehicle front wheel steering angle; the braking system 32 may receive the brake percentage command and perform the corresponding braking; the powertrain 33 may receive an engine torque command or a vehicle speed command and output a corresponding engine torque or vehicle speed.

As shown in fig. 2, the actual vehicle is abstracted into a rectangular vehicle model, which conforms to the ackermann steering principle. The vehicle has a length L and a width W. The position of the vehicle is expressed in terms of the position (x, y) of the actual vehicle rear axle center. The distance from the center of the rear axle to the tail of the vehicle is L_rThe distance from the center of the front axle to the head of the vehicle is L_fThe distance between the front and rear shafts is L_m. The central speed of the rear axle of the vehicle is v, and the rotation angle of the front wheel of the vehicle is vThe vehicle orientation and the X-axis of the global coordinate system form an included angle theta.

As shown in FIG. 3, a typical vertical parking space has vehicles parked on both sides, which are abstracted as a quadrilateral P₁And P₂And four corner points { P thereof are measured by a perception system_j，k|j＝1，2；k＝1，2，3，4}。The target state of the vehicle is set by the sensing system after identifying the type of the parking space, and generally, the vehicle speed is 0 in the target state, namely v_fWhen the vehicle is oriented in the same direction as the parking space, i.e. θ_f＝90°。The initial state of the vehicle is the state when the vehicle starts to apply the present invention, and generally, the vehicle speed is 0, i.e., v_z＝0。For the ith state of the vehicle in the planned path, its four corner points are denoted by { C_i，1，C_i，2，C_i，3，C_i，4Represents it.

The front wheel steering four-wheel vehicle model conforming to the Ackerman steering principle has a steering center when steering and is positioned on a rear axle extension line. At low speeds, the slip of the tire can be neglected and the vehicle kinematic differential equation can be expressed as:

vehicle state variable composed ofDenoted by (a, ω), the control variable. The vehicle state variables and control variables are continuous over time t, and are sampled over a series of times to form a series of states of the vehicle. Segmenting the vehicle state variable and the control variable, initially setting the number of segments to be N, for example, N is 10, each segment comprises 5 sampling points at equal intervals, each segment has the time length of 4h, and the variables x, y, v, theta,In 5 sampling points of each section after omega dispersion, the sampling points at two ends are shared with adjacent sections, and each section after variable a dispersion comprises 5 independent sampling points; the variables to be optimized are obtained as follows:

the differential at each sample point can be expressed as a function of 5 sample points on the segment using lagrange interpolation:

$\left\{\begin{array}{c}{s^{\′}}_{n,0}=\frac{1}{12h}(-25\·s_{n,0}+48\·s_{n,1}-36\·s_{n,2}+16\·s_{n,3}-3\·s_{n,4})\\ {s^{\′}}_{n,1}=\frac{1}{12h}(-3\·s_{n,0}-10\·s_{n,1}+18\·s_{n,2}-6\·s_{n,3}+\·s_{n,4})\\ {s^{\′}}_{n,2}=\frac{1}{12h}(s_{n,0}-8\·s_{n,1}+8\·s_{n,3}-s_{n,4})\\ {s^{\′}}_{n,3}=\frac{1}{12h}(-s_{n,0}+6\·s_{n,1}-18\·s_{n,2}+10\·s_{n,3}+3\·s_{n,4})\\ {s^{\′}}_{n,4}=\frac{1}{12h}(3\·s_{n,0}-16\·s_{n,1}+36\·s_{n,2}-48\·s_{n,3}+25\·s_{n,4})\\ n=0,1,\mathrm{...},N-1\end{array}\right.---\left(2\right)$

wherein s is_n，mN is 0, 1, …, N-1; m is 0, 1, …, 4 is a state variable s, i.e.At time t_n，mValue of (2), S'_n，mFor a state variable s at a time t_n,mDerivative of (a), t_n，mH (4 × n + m) denotes the time instant at the mth sampling point within the nth segment.

Are all functions of time t, i.e. can be expressed asIn the form of (1). Simultaneous formulas (1) and (2), wherein the form in formula (1) is as followsThe differential equation of (a) is converted into the following algebraic equation, forming the equality constraint of the variable to be optimized:

s′_n，m*h-f(t_n，m)*h＝0，(n＝0，1…，N-1；m＝0，1…，4) (3)

there is a limit in the movement of the vehicle, i.e. its front wheel angle and its front wheel angular velocity have a maximum in both forward and reverse directions. Also for safety reasons during parking, the range of motion (x, y) of the vehicle and the speed and acceleration of the vehicle should be limited. The time step h, the number of segments N and the number of sampling points 5 together determine the parking time, and the parking time is usually not too long, so the time step h is limited. Meanwhile, the first and last states of the vehicle in the planning result should be equal to the initial state and the target state of the vehicle, respectively. In summary, the boundary constraints of the variables to be optimized are as follows:

wherein h is_maxMaximum limit value, x, representing a time step_lb、y_lb、v_lb、a_lb、θ_lb、ω_lbRespectively represent variables x, y, v, a, theta,Lower limit of ω, x_ub、y_ub、v_ub、a_ub、θ_ub、ω_ubRespectively represent variables x, y, v, a, theta,The upper limit value of ω.

It is most important that the vehicle does not collide with obstacles during parking along the planned path. Using the area method, it can be determined whether a point is located within a quadrilateral: when the point is positioned in the quadrangle, the area of four triangles formed by the point and four sides of the quadrangle is equal to the area of the quadrangle; when a point is located outside the quadrilateral, the area of four triangles formed by the point and four sides of the quadrilateral is larger than the area of the quadrilateral. If for each state of the vehicle, { P_j，k1, 2; k is outside its four-sided polygon and for each obstacle, the four corner points { C of each state of the vehicle_i，1，C_i，2，C_i，3，C_i，4Are outside its four-sided polygon, it can be determined that each state of the vehicle is safe and collision-free. Thus, the inequality constraints for the variables to be optimized are:

$\left\{\begin{array}{c}S(C_i,P_{j,k})>\mathrm{\α}*SA,(i=0,1,\mathrm{...},4N;j=1,2;k=1,2,3,4)\\ S(P_j,C_{i,k})>\mathrm{\α}*{\mathrm{SP}}_j,(i=0,1,\mathrm{...},4N;j=1,2;k=1,2,3,4)\end{array}\right.---\left(5\right)$

wherein, S (C)_i，P_j，k) Represents P_j，kAnd a quadrangle C_iThe sum of the areas of the four triangles formed, SA, representing the quadrilateral C_iArea of (d), S (P)_j，C_i，k) Is represented by C_i，kAnd a quadrangle P_jFour three being formedSum of angular areas, SP_jRepresenting a quadrilateral P_jα is a safety factor greater than 1, e.g., the greater the 1.05, α, the greater the safe separation of the vehicle from the barrier.

Setting the optimization target to be the shortest time, namely the target function is as follows:

T_f＝N*4h (6)

the vehicle has a limited moving range during parking, for example, the vehicle cannot excessively invade another lane to prevent other vehicles from passing through, and meanwhile, the vehicle brings safety hazards to the vehicle. In addition, the two sides of the road may be a space restricted area such as a wall, so that the parking path planning should be applied with a motion range constraint. During parking, the four corner points of the vehicle may not exceed the limited motion range, so the motion range of the vehicle is constrained as follows:

$\left\{\begin{array}{c}{x}_{lb}\≤x_{C_{i,k}}\≤x_{ub},(i=1,\mathrm{2...},4N-1;k=1,2,3,4)\\ y_{lb}\≤y_{C_{i,k}}\≤y_{ub},(i=1,2,\mathrm{...},4N-1;k=1,2,3,4)\end{array}\right.---\left(7\right)$

wherein,x-axis coordinates of a kth corner point at an i-th state of the vehicle,the Y-axis coordinate of the kth corner point at the i-th state of the vehicle.Andcan be calculated from the vehicle state parameters and the geometric parameters of the vehicle.

Solving the above obtained constrained nonlinear programming problem using a nonlinear programming solver, such as IPOPT, SNOPT:

when the parking environment is too harsh, the solver cannot obtain a solution which meets the constraint, and at the moment, the parking path planning is judged to fail. Otherwise, the result obtained by the solution is 4N +1 vehicle states representing the shortest parking path, and the values of the vehicle state variables and the control variables at any time can be obtained by fitting by using a segmented Lagrange interpolation method. In order to prevent the time step h from being too large to cause the collision between the two planned vehicle states to be detected, 0.01 second is used as a refined time step to obtain a more refined vehicle state sequence. And detecting whether each vehicle state in the refined vehicle state sequence is collided, and if not, judging that the parking path planning is successful. Otherwise, increasing the number of segments, for example, making the number of segments 2N, and performing path planning again. And after repeating for 3 times, if the collision still exists, judging that the parking path planning is failed.

And finally, when the parking path is planned successfully, the parking controller controls the power system, the braking system and the steering system in real time to track the planned parking path.

The invention is applicable to various parking scenes, and fig. 4 and 5 respectively show parking paths obtained by applying the invention under the conditions of oblique parking and parallel parking of the vehicle.

It should be noted that the present invention is not limited to other vehicles as the obstacle around the parking space, and may also be other obstacles, such as a parking space ground lock, which is also applicable to the application scenario of the present invention. It should be noted that the number of obstacles around the parking space is not limited to 2, and other numbers of obstacles are also applicable to the application scenario of the present invention.

The above-mentioned embodiments are merely illustrative of the preferred embodiments of the present invention, and do not limit the scope of the present invention, and various modifications and improvements made to the technical solution of the present invention by those skilled in the art without departing from the spirit of the present invention should fall within the protection scope defined by the claims of the present invention.

Claims (10)

1. A vehicle autonomous parking path planning method for a plurality of parking scenarios for an autonomous parking system to automatically park a vehicle in an available parking space upon detection of the parking space, comprising the steps of:

(1) detecting target parking space information and determining a parking scene;

(2) determining an initial state and a target state of a vehicle to be parked according to the target parking space information and the parking scene;

(3) establishing a vehicle kinematic differential equation based on an Ackerman model of the front-wheel steering four-wheel vehicle;

(4) segmenting the vehicle state variable and the control variable, and carrying out equidistant sampling on each segment according to a certain time step length to obtain a variable to be optimized;

(5) expressing the differential of each sampling point of the vehicle state variable as a function of each sampling point on a section where the sampling point is positioned by adopting a Lagrange interpolation method, and simultaneously connecting the function and a vehicle kinematic differential equation to convert the vehicle kinematic differential equation into an algebraic equation to form an equality constraint of the variable to be optimized;

(6) forming boundary constraints of variables to be optimized according to physical limitations of vehicle motion and safety requirements of parking;

(7) formulating collision avoidance requirements according to obstacles around a target parking space to form inequality constraints of variables to be optimized;

(8) forming a motion range constraint of the vehicle to be parked according to the motion range limit of the vehicle in the parking process;

(9) determining an optimization target and establishing an objective function;

(10) and obtaining the optimal solution of the parking path by adopting a nonlinear programming solver.

2. The method for vehicle autonomous parking path planning for multiple parking scenarios of claim 1, further comprising the steps of:

and (3) fitting the vehicle state variables in the optimal solution of the parking path by adopting a Lagrange interpolation method, sampling the fitted vehicle state variables by using a refined time step length to obtain a refined vehicle state sequence, detecting whether each vehicle state in the refined vehicle state sequence is collided, if so, increasing the number of segments, and repeating the steps (4) to (10) to plan the path again.

3. The method for planning the autonomous parking path of the vehicle for multiple parking scenes according to claim 1, wherein in the step (1), the target parking space information includes the orientation, position, length and width of the target parking space and the position of the obstacle around the target parking space; the parking scenarios include vertical parking, diagonal parking, and parallel parking.

4. The vehicle autonomous parking path planning method for multiple parking scenarios according to claim 1, characterized in that in step (3), the vehicle kinematic differential equation is:

wherein X represents the abscissa of the center of the rear axle of the vehicle in a Cartesian coordinate system, y represents the ordinate of the center of the rear axle of the vehicle in the Cartesian coordinate system, v represents the moving speed of the center of the rear axle of the vehicle, theta represents the angle between the vehicle heading and the X axis of the Cartesian coordinate system,representing the angle of rotation of the front wheels of the vehicle, a representing the acceleration of the centre of the rear axle of the vehicle, ω representing the angular velocity of the angle of rotation of the front wheels of the vehicle, L_mRepresenting the distance between the front axle and the rear axle of the vehicle;

in the step (4), the vehicle state variables are x, y, v, theta,The vehicle control variable is a and omega, the vehicle state variable and the control variable are segmented, the number of the segments is N, each segment comprises M sampling points at equal intervals, the time length of each segment is (M-1) h, and the variables x, y, v, theta and delta are measured by a computer,In the M sampling points of each segment after ω dispersion, the sampling points at both ends are shared with the adjacent segments, and each segment after variable a dispersion contains M independent sampling points, then the variable to be optimized is:

wherein h represents a time step; x is the number of_i(M-1) N represents the value of the ith sampling point after the variable x is discretized; y is_i(M-1) N represents the value of the ith sampling point after the variable y is discretized; v. of_i(M-1) N represents the value of the ith sample point after the dispersion of the variable v; theta_i(M-1) N represents the value of the ith sample point after the variation θ is discretized,representing variablesThe value of the ith sample point after dispersion; omega_i(M-1) N represents the value of the ith sampling point after the variable ω is discretized; a is_pAnd p is 0, 1, and MN-1 represents the value of the p-th sampling point after the variable a is dispersed.

5. The method for vehicle autonomous parking path planning for multiple parking scenarios of claim 4 wherein Lagrangian interpolation is used to interpolate the vehicle state variables x, y, v, θ,The differential at each sample point is expressed as a function of the M sample points over the segment in which the sample point lies:

$s_{n,m}^{\′}=\frac{1}{h}\underset{m=0}{\overset{M-1}{\Σ}}{\mathrm{\γ}}_{n,m}*s_{n,m},n=0,1,...,N-1;m=0,1,...,M-1$

and combining the function and the vehicle kinematic differential equation to convert the vehicle kinematic differential equation into an algebraic equation to form an equation constraint of a variable to be optimized:

S′_n，m*h-f(t_n，m)*h＝0，n＝0，1，…，N-1；m＝0，1，...，M-1

wherein s represents a vehicle state variable x, y, v, θ,s_n，mRepresenting vehicle state variables at time t_n，mValue of (a), gamma_n，mRepresenting vehicle state variables at time t_n，mCoefficient of (d), s'_n，mRepresenting vehicle state variables at time t_n，mDerivative of (a), t_n，m＝h*[(M-1)*n+m]Indicating the time instant at the mth sampling point within the nth segment.

6. The method for vehicle autonomous parking path planning for multiple parking scenarios according to claim 4, wherein the boundary constraints of the variables to be optimized are:

wherein h is_maxMaximum limit value, x, representing a time step_lb、y_lb、v_lb、a_lb、θ_lb、ω_lbRespectively represent variables x, y, v, a, theta,Lower limit of ω, x_ub、y_ub、v_ub、a_ub、θ_ub、ω_ubRespectively represent variables x, y, v, a, theta,An upper limit value of ω;indicating an initial state of the vehicle to be parked,indicating a target state of the vehicle to be parked; x is the number of_i、y_i、v_i、θ_i、ω_iRespectively represent variables x, y, v, theta,Value of ith sample point after ω dispersion, a_pRepresenting the value of the p-th sample point after the dispersion of the variable a.

7. The method for vehicle autonomous parking path planning for multiple parking scenarios according to claim 4, wherein the inequality constraints of the variables to be optimized are:

$\left\{\begin{array}{cc}S(C_i,P_{j,k})>\mathrm{\α}*SA,& i=0,1,...,(M-1)N;j=1,2,...,J;k=1,2,3,4\\ S(P_j,C_{i,k})>\mathrm{\α}*{\mathrm{SP}}_j,& i=0,1,...,(M-1)N;j=1,2,...,J;k=1,2,3,4\end{array}\right.$

wherein, C_iRepresenting a quadrilateral into which the vehicle to be parked is abstracted when it is in the i-th state, P_jRepresenting a quadrilateral into which the jth obstacle is abstracted, J representing the number of obstacles, P_j，kRepresenting a quadrilateral P_jOf the kth corner point, C_i，kRepresents a quadrangle C_iThe kth corner of (2), S (C)_i，P_j，k) Represents P_j，kAnd a quadrangle C_iThe sum of the areas of the four triangles formed, SA, representing the quadrilateral C_iArea of (d), S (P)_j，C_i，k) Is represented by C_i，kAnd a quadrangle P_jSum of areas of four triangles formed, SP_jRepresenting a quadrilateral P_jα is a safety factor greater than 1.

8. The vehicle autonomous parking path planning method for multiple parking scenarios according to claim 4, characterized in that the range of motion of the vehicle to be parked is constrained as follows:

$\left\{\begin{array}{cc}{x}_{lb}\≤x_{C_{i,k}}\≤x_{ub},& i=1,2,...,\left(M-1\right)N-1;k=1,2,3,4\\ y_{lb}\≤y_{C_{i,k}}\≤y_{ub},& i=1,2,...,\left(M-1\right)N-1;k=1,2,3,4\end{array}\right.$

wherein,x-axis coordinates representing a kth corner point of a quadrangle into which the vehicle to be parked is abstracted when it is in the ith state,to indicate a waitA Y-axis coordinate of a k-th corner point of the quadrangle abstracted when the parked vehicle is in the i-th state; x is the number of_lb、x_ubRespectively representing the lower and upper limits of a vehicle state variable x, y_lb、y_ubRespectively representing the lower limit value and the upper limit value of the vehicle state variable y.

9. The method of claim 4, wherein the objective function is:

T_f＝N*(M-1)h

wherein, T_fRepresenting a path planning time of the vehicle to be parked;

the optimization objective is to minimize the time, i.e., min T_f。

10. The method for vehicle autonomous parking path planning for multiple parking scenarios of claim 4 wherein M is an integer and 4 ≦ M ≦ 8.