Trajectory Optimization Method, Device, Terminal and Storage Medium for Emergency Obstacle Avoidance in Autonomous Driving of Vehicles

Abstract

The present application discloses a trajectory optimization method, device, terminal, and storage medium for emergency obstacle avoidance in vehicle autonomous driving. The method includes receiving preset basic parameters, including the starting point of the vehicle driving trajectory, the ending point of the trajectory, and the speed planning equation; establishing a trajectory equation of the trajectory, setting multiple trajectory calculation points between the starting point and the ending point according to the trajectory equation, and calculating the longitudinal speed and longitudinal acceleration of each trajectory calculation point according to the speed planning equation; calculating the comprehensive acceleration of each trajectory calculation point by combining the trajectory equation and the speed and longitudinal acceleration of each trajectory calculation point, and screening to obtain the maximum comprehensive acceleration; optimizing and iterating the coefficients of each trajectory polynomial of the trajectory equation by combining the maximum comprehensive acceleration to obtain the optimal polynomial coefficients, and performing trajectory planning by combining the optimal polynomial coefficients. The method of the present application minimizes the maximum comprehensive acceleration of the planned trajectory, thereby improving the safety level during the emergency obstacle avoidance process.

Description

Technical Field

[0001] This application relates to the field of automotive advanced driver assistance systems (ADAS), and particularly to a trajectory optimization method, device, terminal, and storage medium for emergency obstacle avoidance in automotive autonomous driving. Background Art

[0002] There are two tasks in the trajectory planning of ADAS: one is speed planning, that is, reasonably arranging the speed of the vehicle, which can be uniform or variable; the other is to reasonably arrange the curvature of the trajectory. Because the greater the curvature, the greater the steering angle of the vehicle, the greater the lateral acceleration, that is, the greater the centrifugal force. When the centrifugal force reaches a certain level, the resultant force of the centrifugal force and the longitudinal force will cause slippage between the tire and the ground, resulting in vehicle out of control. Therefore, an important task in trajectory planning is to ensure a reasonable trajectory curvature so that when the vehicle travels along the planned trajectory, the vector sum (composite acceleration) of the lateral acceleration and the longitudinal acceleration is minimized and kept within a safe range.

[0003] Currently, the trajectory planning of ADAS is calculated according to the traditional fifth-degree polynomial. Under the constraints of the positions, speed directions, and curvatures at the starting point and the ending point, the coefficients of the fifth-degree polynomial are completely solvable, that is, the curvature of the trajectory does not change with the change of speed planning. In fact, when performing different speed plans, the speeds of points on the trajectory will change. To ensure that the lateral acceleration (or composite acceleration) is within a certain safe range, for points with a high speed, the curvature should be smaller, and for points with a low speed, the curvature can be moderately larger. However, the calculation method of the fifth-degree polynomial cannot achieve curvature optimization under different speed plans, and thus cannot control the maximum composite acceleration when the vehicle travels along the planned trajectory to the minimum level.

[0004] Therefore, how to control the maximum composite acceleration when the vehicle travels along the planned trajectory to the minimum level is an urgent problem to be solved in this field. Summary of the Invention

[0005] To solve the problem that the existing obstacle avoidance trajectory planning technology cannot reasonably design the trajectory curvature according to speed planning, this application represents the trajectory curve based on a multi-degree polynomial of the trajectory and simultaneously uses optimization calculation to minimize the composite acceleration when the vehicle travels along the trajectory, thereby improving the safety of the emergency obstacle avoidance function.

[0006] In a first aspect, this application provides a trajectory optimization method for emergency obstacle avoidance in automotive autonomous driving, adopting the following technical solutions:

[0007] A trajectory optimization method for emergency obstacle avoidance in automotive autonomous driving includes the following steps:

[0008] Receive preset basic parameters, where the basic parameters include the starting point of the trajectory of the vehicle's travel and the starting point parameters of the starting point, and also include the ending point of the trajectory and the ending point parameters of the ending point, and receive the pre-established speed planning equation of the trajectory;

[0009] Establish the trajectory equation of the trajectory, set a plurality of trajectory calculation points between the starting point and the ending point according to the trajectory equation, and calculate the longitudinal speed and longitudinal acceleration of each trajectory calculation point according to the speed planning equation;

[0010] Combine the trajectory equation and the speed and longitudinal acceleration of each trajectory calculation point to calculate the comprehensive acceleration of each trajectory calculation point and obtain a set of comprehensive accelerations, and screen the maximum comprehensive acceleration from the set of comprehensive accelerations;

[0011] The trajectory equation includes a plurality of trajectory polynomial coefficients. Optimize and iterate each trajectory polynomial coefficient in combination with the maximum comprehensive acceleration to obtain the optimal polynomial coefficients, and update the trajectory equation in combination with the optimal polynomial coefficients. The updated trajectory equation is used for trajectory planning.

[0012] By adopting the above technical means, a detailed trajectory equation and a speed planning equation are established, and in combination with the data of actual trajectory calculation points, highly accurate control of the vehicle's motion state is achieved; at the same time, by performing optimization calculations on the trajectory polynomial coefficients of the trajectory equation, the maximum comprehensive acceleration at each point during the vehicle's emergency obstacle avoidance process is minimized, reducing the risk of tire skidding caused by excessive lateral acceleration and improving driving safety.

[0013] Preferably, the steps of establishing the trajectory equation of the trajectory and setting a plurality of trajectory calculation points between the starting point and the ending point according to the trajectory equation specifically include the following steps:

[0014] Establish the trajectory equation of the trajectory. The trajectory is the driving curve of the vehicle from the starting point to the ending point. The trajectory equation is a polynomial equation of multiple degrees, and the trajectory equation includes a plurality of trajectory polynomial coefficients;

[0015] Set a plurality of trajectory calculation points between the starting point and the ending point according to the trajectory equation. All the trajectory calculation points are located on the trajectory, and calculate the first derivative and the second derivative of each trajectory calculation point respectively.

[0016] By adopting the above technical means, the first derivative and the second derivative of each calculation point on the trajectory can be accurately calculated, and then the speed and acceleration changes of the vehicle at these calculation points can be accurately reflected, which helps to more finely control the driving path of the vehicle, ensure that the vehicle always stays within the safe comprehensive acceleration range during the emergency obstacle avoidance process, and improve the safety and reliability of the obstacle avoidance process.

[0017] Preferably, the trajectory equation is

[0018] Y = A×X M 6 + B×X M 5 + C×X M 4 + D×X M 3 + E×X M 2 + F×X M + G;

[0019] Wherein, X M is the longitudinal coordinate of any trajectory calculation point on the trajectory between the starting point and the ending point, Y is the transverse coordinate of the trajectory calculation point corresponding to X M , and A, B, C, D, E, F, G are the trajectory polynomial coefficients of the trajectory equation;

[0020] The calculation formulas for the first derivative and the second derivative of the trajectory calculation point are

[0021] Y’ = 6A×X M 5 + 5B×X M 4 + 4C×X M 3 + 3D×X M 2 + 2E×X M + F;

[0022] Y’’ = 30A×X M 4 + 20B×X M 3 + 12C×X M 2 + 6D×X M + 2E;

[0023] Wherein, Y’ is the first derivative of the trajectory calculation point corresponding to X M , and Y’’ is the second derivative of the trajectory calculation point corresponding to X M .

[0024] By adopting the above technical means, the curvature of each trajectory calculation point can be accurately obtained by calculating the first and second derivatives of the trajectory equation, thereby better controlling the steering angle and lateral acceleration of the vehicle. Moreover, the detailed derivative calculation formula provides solid data support for subsequent optimization calculations, ensuring that the dynamic performance index of each point can be accurately evaluated during the optimization process.

[0025] Preferably, calculating the longitudinal speed and longitudinal acceleration of each of the trajectory calculation points according to the speed planning equation specifically includes the following steps:

[0026] The speed planning equation includes

[0027] V M =Q×t 3 +R×t 2 +U×t+W;

[0028] a xM =3×Q×t 2 +2×R×t+U;

[0029] where t is any moment after the vehicle starts to avoid obstacles, and Q, R, U, and W are speed planning polynomial coefficients. V M is the longitudinal speed of the corresponding trajectory calculation point at time t, and a xM is the longitudinal acceleration of the corresponding trajectory calculation point at time t;

[0030] Calculating the longitudinal speed and longitudinal acceleration of each of the trajectory calculation points in the trajectory according to the speed planning equation.

[0031] By adopting the above technical means, by introducing the speed planning equation, the longitudinal speed and longitudinal acceleration of each trajectory calculation point are accurately calculated, providing an accurate data basis for the subsequent calculation of the combined acceleration.

[0032] Preferably, calculating the combined acceleration of each of the trajectory calculation points by combining the trajectory equation and the speed and longitudinal acceleration of each of the trajectory calculation points to obtain a combined acceleration set, and screening the maximum combined acceleration from the combined acceleration set specifically includes the following steps:

[0033] Calculating the curvature K of the current trajectory calculation point according to the first derivative and second derivative of any of the trajectory calculation points M ;

[0034] Calculating the lateral acceleration a yM of the current trajectory calculation point according to the longitudinal speed and the curvature of the current trajectory calculation point, and the calculation formula is

[0035] ayM =V M 2 ×K M ;

[0036] Calculate the lateral acceleration and longitudinal acceleration of the points based on the current trajectory, and obtain the comprehensive acceleration a of the current trajectory calculation point SM , and the calculation formula is

[0037] a SM =(a yM 2 +a xM 2 ) 1 / 2 ;

[0038] Calculate the comprehensive acceleration of each trajectory calculation point and summarize to obtain a comprehensive acceleration set, and screen the maximum comprehensive acceleration from the comprehensive acceleration set. The calculation formula is

[0039] a SMAX =Max(a SM );

[0040] where a SMAX is the maximum comprehensive acceleration among the comprehensive accelerations corresponding to all the trajectory calculation points

[0041] By adopting the above technical means, the curvature, lateral acceleration and comprehensive acceleration of each trajectory calculation point can be accurately calculated, and the maximum comprehensive acceleration can be screened out from them, which helps to ensure that the maximum comprehensive acceleration of the vehicle along the planned trajectory is always within the safe range during the emergency obstacle avoidance process, improving driving safety

[0042] Preferably, optimize and iterate the coefficients of each trajectory polynomial by combining the maximum comprehensive acceleration to obtain the optimal polynomial coefficients, which specifically include the following steps

[0043] The basic parameters also include the coefficient threshold of the trajectory polynomial coefficient A, and determine the initial value of the trajectory polynomial coefficient A according to the coefficient threshold

[0044] Optimize and iterate the coefficients of each trajectory polynomial according to the preset constraint rules and the initial value. The process of the optimization iteration includes

[0045] Obtain the relational expressions of the trajectory polynomial coefficients B, trajectory polynomial coefficient C, trajectory polynomial coefficient D, trajectory polynomial coefficient E, trajectory polynomial coefficient F and trajectory polynomial coefficient G according to the constraint rules and the initial value, establish the optimization objective function, and obtain the optimal polynomial coefficients of the trajectory equation according to the optimization objective function

[0046] By adopting the above technical means, through introducing the initial value of the trajectory polynomial coefficient A and multiple constraint rules, and optimizing and iterating each trajectory polynomial coefficient, more accurate trajectory planning is achieved.

[0047] Preferably, the starting point parameters include the starting point coordinates, the starting point curvature, and the initial heading angle, and the ending point parameters include the ending point coordinates, the ending point curvature, and the ending point heading angle;

[0048] The constraint rules include multiple constraint conditions,

[0049] Among them, the first constraint condition is that the curve represented by the trajectory equation passes through the starting point coordinates;

[0050] The second constraint condition is that the curve slope of the starting point coordinates is the tangent function of the initial heading angle;

[0051] The third constraint condition is that the curve represented by the trajectory equation passes through the ending point coordinates;

[0052] The fourth constraint condition is that the curve slope of the ending point coordinates is the tangent function of the ending point heading angle;

[0053] According to the first constraint condition, the second constraint condition, the third constraint condition, and the fourth constraint condition, combined with the curvatures of the starting point and the ending point, the relational expressions of the trajectory polynomial coefficients B, C, D, E, F, and G are obtained.

[0054] By adopting the above technical means, through specifying multiple constraint conditions, it is ensured that the trajectory equation can accurately reflect the actual driving requirements, and at the same time, the relational expression of the trajectory polynomial coefficients is obtained, realizing the optimization of the coefficients. The optimized trajectory equation helps to ensure that the comprehensive acceleration of the vehicle always remains within the safe range under different speed plans, significantly improving the safety and reliability of emergency obstacle avoidance.

[0055] In a second aspect, the present application provides a trajectory optimization device for emergency obstacle avoidance in vehicle autonomous driving, adopting the following technical solution:

[0056] A trajectory optimization device for emergency obstacle avoidance in vehicle autonomous driving includes the following modules:

[0057] A basic parameter acquisition module, configured to receive preset basic parameters, the basic parameters including the starting point of the trajectory of the vehicle driving and the starting point parameters of the starting point, and further including the ending point of the trajectory and the ending point parameters of the ending point, and receive the pre-established speed planning equation of the trajectory;

[0058] A speed planning module, configured to establish a trajectory equation of the trajectory, set a plurality of trajectory calculation points between the starting point and the ending point according to the trajectory equation, and calculate the longitudinal speed and longitudinal acceleration of each of the trajectory calculation points according to the speed planning equation;

[0059] A maximum comprehensive acceleration calculation module, configured to calculate the comprehensive acceleration of each of the trajectory calculation points in combination with the trajectory equation and the speed and the longitudinal acceleration of each of the trajectory calculation points to obtain a comprehensive acceleration set, and screen out the maximum comprehensive acceleration from the comprehensive acceleration set;

[0060] A coefficient optimization module, wherein the trajectory equation includes a plurality of trajectory polynomial coefficients, and is configured to perform optimization iteration on each of the trajectory polynomial coefficients in combination with the maximum comprehensive acceleration to obtain optimal polynomial coefficients, and update the trajectory equation in combination with the optimal polynomial coefficients, and the updated trajectory equation is used for trajectory planning.

[0061] By adopting the above technical means, a complete trajectory optimization system for emergency obstacle avoidance is built, which provides necessary software technical support for the precise operation of emergency obstacle avoidance in vehicle autonomous driving, significantly improves the safety and stability of the obstacle avoidance process, effectively reduces the risk of vehicle out of control caused by high lateral acceleration, and meets the requirements of technological progress.

[0062] In a third aspect, the present application provides an intelligent terminal, adopting the following technical solution:

[0063] An intelligent terminal includes a memory and a processor. At least one instruction, at least one program, a code set or an instruction set is stored in the memory, and the at least one instruction, at least one program, the code set or the instruction set is loaded and executed by the processor to implement the trajectory optimization method for emergency obstacle avoidance in vehicle autonomous driving as described above.

[0064] In a fourth aspect, the present application provides a computer-readable storage medium, adopting the following technical solution:

[0065] A computer-readable storage medium stores at least one instruction, at least one program, a code set or an instruction set, and the at least one instruction, at least one program, the code set or the instruction set is loaded and executed by a processor to implement the trajectory optimization method for emergency obstacle avoidance in vehicle autonomous driving as described above.

[0066] In summary, the present application at least includes the following beneficial effects:

[0067] (1) This application uses a multi - degree polynomial equation for trajectory planning and combines an optimization calculation method to make the trajectory curvature of the vehicle reach the optimal state under different speed plans, thereby reducing the maximum combined acceleration and ensuring that the vehicle remains within a safe range during an emergency obstacle avoidance process.

[0068] (2) This application can automatically adjust the trajectory curvature under different speed planning conditions, enabling the vehicle to reduce the lateral acceleration when driving at high speed and appropriately increase the curvature when driving at low speed, achieving a more flexible and efficient path planning.

[0069] (3) By introducing multiple constraint conditions, this application optimizes the calculation of the polynomial coefficients of the trajectory to ensure that the trajectory planning meets the requirements of various parameters, improves the stability and robustness of the system, and reduces the probability of unexpected situations. BRIEF DESCRIPTION OF THE DRAWINGS

[0070] Figure 1 is one of the flow schematic diagrams of the trajectory optimization method for emergency obstacle avoidance in vehicle autonomous driving according to an embodiment of this application;

[0071] Figure 2 is another flow schematic diagram of the trajectory optimization method for emergency obstacle avoidance in vehicle autonomous driving according to an embodiment of this application;

[0072] Figure 3 is the path curve schematic diagram of the trajectory optimization method for emergency obstacle avoidance in vehicle autonomous driving according to an embodiment of this application;

[0073] Figure 4 is the path and combined acceleration curve diagram of the comparison between optimization and non - optimization in Embodiment 1 of this application;

[0074] Figure 5 is the longitudinal speed and longitudinal acceleration curve diagram of Embodiment 1 of this application;

[0075] Figure 6 is the path and combined acceleration curve diagram of the comparison between optimization and non - optimization in Embodiment 2 of this application;

[0076] Figure 7 is the longitudinal speed and longitudinal acceleration curve diagram of Embodiment 2 of this application;

[0077] Figure 8 is the structural diagram of the trajectory optimization for emergency obstacle avoidance in vehicle autonomous driving according to an embodiment of this application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0078] This application provides a trajectory optimization method, device, terminal, and storage medium for emergency obstacle avoidance in vehicle autonomous driving. To make the objectives, technical solutions, and advantages of this application clearer, the following will further elaborate on the embodiments of this application in detail.

[0079] The following further describes in detail an embodiment of a trajectory optimization method for emergency obstacle avoidance in autonomous driving of a vehicle in conjunction with the accompanying drawings of the specification.

[0080] A trajectory optimization method for emergency obstacle avoidance in autonomous driving of a vehicle according to the present application, as Figure 1 and Figure 2 shown, includes the following steps:

[0081] S1. Receive preset basic parameters, where the basic parameters include the starting point of the trajectory of the vehicle driving and the starting point parameters of the starting point, and also include the ending point of the trajectory and the ending point parameters of the ending point, and receive the pre-established speed planning equation of the trajectory.

[0082] S11. Receive preset basic parameters.

[0083] Specifically, the starting point parameters include the starting point coordinates, the starting point curvature, and the initial heading angle.

[0084] The ending point parameters include the ending point coordinates, the ending point curvature, and the ending point heading angle;

[0085] The basic parameters also include the coefficient threshold of the coefficient A of the trajectory polynomial of the trajectory equation to be established.

[0086] S12. Receive the pre-established speed planning equation.

[0087] The speed planning equation is used for speed planning of the vehicle. In this embodiment, the speed planning equation is obtained from the upstream system and established in the upstream system.

[0088] The speed planning equation in this embodiment is a cubic polynomial.

[0089] In this embodiment, the speed planning equation includes,

[0090] V M = Q × t 3 + R × t 2 + U × t + W;

[0091] a xM = 3 × Q × t 2 + 2 × R × t + U;

[0092] where t is any moment after the vehicle starts to avoid obstacles, Q, R, U, and W are speed planning polynomial coefficients, V M is the longitudinal speed of the trajectory calculation point corresponding to the moment t, and a xM is the longitudinal acceleration of the trajectory calculation point corresponding to the moment t.

[0093] The basic parameters also include the time to collision TTC and the speed planning end time Ts.

[0094] In this embodiment, the starting point is set as O and the ending point is set as S;

[0095] The coordinates of the starting point O are (0.0, 0.0), the coordinates of the ending point S are (Lg, Hg), the speed planning curve coefficients are Q, R, U, W, the speed planning end time is Ts, the time to collision is TTC (Time to Collision), and the starting heading angle of the starting point O is α O and the starting curvature is K O ; the ending heading angle of the ending point S is α S and the ending curvature is K S .

[0096] Among them, TTC (Time to Collision) is the input of the previous module, indicating the time when the vehicle collides with an obstacle. When the previous module discovers a dangerous obstacle in front of the vehicle, it will calculate how long it will take for the vehicle to collide with the obstacle without taking evasive measures. This time is TTC. After obtaining TTC, a speed planning equation is given within the time range of 0 - TTC. Then, the speed planning polynomial coefficients and TTC are input into the next module for optimizing the obstacle avoidance trajectory. Since the problem discussed in this application has nothing to do with the calculation of TTC, the content of this module will not be elaborated here.

[0097] For this embodiment, TTC is the end time of the speed planning equation, and Ts is the actual speed planning end time. Therefore, Ts can be equal to TTC. However, if Ts > TTC, the vehicle speed for the time exceeding TTC remains constant. For example, if TTC is 4 seconds, the vehicle speed for the time less than 4 seconds is calculated according to the speed planning equation, and the vehicle speed for the time greater than 4 seconds is taken as the vehicle speed calculated according to the speed planning equation at the moment of 4 seconds.

[0098] S2. Establish the trajectory equation of the trajectory, set multiple trajectory calculation points between the starting point and the ending point according to the trajectory equation, establish the speed planning equation of the trajectory, and calculate the longitudinal speed and longitudinal acceleration of each trajectory calculation point according to the speed planning equation. The specific steps are as follows:

[0099] S21. Establish the trajectory equation of the trajectory. The trajectory is the driving curve of the vehicle from the starting point to the ending point.

[0100] As Figure 3 shown in the path curve graph, the coordinates of the starting point O are (L0, H0), the starting heading angle is α O , the coordinates of the ending point S are (Lg, Hg), and the ending heading angle is α S, M(X, Y) is an arbitrary point on the curve, and the longitudinal velocity of M is V M , and the longitudinal acceleration is a xM , and the lateral acceleration is a yM , and Z is the end point of the velocity planning.

[0101] The trajectory equation is a polynomial equation of multiple degrees, and the trajectory equation contains multiple trajectory polynomial coefficients.

[0102] In this embodiment, a trajectory equation is established with a sixth-degree polynomial as an example. In practical applications, a seventh-degree polynomial or even an eighth-degree polynomial can also be used to establish the trajectory equation.

[0103] The trajectory equation in this embodiment is

[0104] Y = A×X M 6 + B×X M 5 + C×X M 4 + D×X M 3 + E×X M 2 + F×X M + G,

[0105] where X M is the longitudinal coordinate of any trajectory calculation point on the trajectory between the starting point and the ending point, Y is the transverse coordinate of the trajectory calculation point corresponding to X M , and A, B, C, D, E, F, G are the trajectory polynomial coefficients of the trajectory equation.

[0106] S22. Set multiple trajectory calculation points between the starting point and the ending point according to the trajectory equation, and all the trajectory calculation points are located on the trajectory.

[0107] The more trajectory calculation points there are between the starting point and the ending point, the higher the accuracy, but the greater the computational amount and the more system resources are consumed. In this embodiment, 100 trajectory calculation points are taken between the starting point and the ending point.

[0108] The number of trajectory calculation points is determined according to the specific application situation, but it is necessary to make the calculation results differ by less than 5%. For example, in a specific implementation, if the calculation result with 1000 trajectory calculation points is α and the calculation result with 100 trajectory calculation points is β, and (α - β) / β < 5%, then 100 trajectory calculation points meet the requirements.

[0109] S23. Calculate the first derivative and the second derivative of each trajectory calculation point respectively.

[0110] In this embodiment, the calculation formulas for the first derivative and the second derivative of the trajectory calculation points are

[0111] Y’ = 6A × X M 5 + 5B × X M 4 + 4C × X M 3 + 3D × X M 2 + 2E × X M + F;

[0112] Y’’ = 30A × X M 4 + 20B × X M 3 + 12C × X M 2 + 6D × X M + 2E;

[0113] Wherein, Y’ is the first derivative of the trajectory calculation point corresponding to X M and Y’’ is the second derivative of the trajectory calculation point corresponding to X M S24. Calculate the longitudinal speed and longitudinal acceleration of each trajectory calculation point in the trajectory according to the obtained speed planning equation.

[0114] S3. Calculate the comprehensive acceleration of each trajectory calculation point and obtain the comprehensive acceleration set by combining the trajectory equation and the speed and longitudinal acceleration of each trajectory calculation point. Screen the maximum comprehensive acceleration from the comprehensive acceleration set. The specific steps are as follows:

[0115] S31. Calculate the curvature K of the current trajectory calculation point according to the first derivative and second derivative of any trajectory calculation point

[0116] In this embodiment, the curvature calculation formula of any trajectory calculation point M(X, Y) on the trajectory is as follows M ,

[0117] K

[0118] K M = Y’’ / (1 + Y’ 2 ) 3 / 2 ;

[0119] Wherein, K M is the curvature of any trajectory calculation point M.

[0120] S32. Calculate the lateral acceleration a of the current trajectory calculation point M according to the longitudinal speed and curvature of the current trajectory calculation point. The calculation formula is as follows yM The calculation formula is

[0121] a yM = V M2 ×K M ;

[0122] where a yM is the lateral acceleration of the current trajectory calculation point M.

[0123] S33. Calculate the combined acceleration a SM of the current trajectory calculation point M based on the lateral acceleration and longitudinal acceleration of the current trajectory calculation point. The calculation formula is

[0124] a SM =(a yM 2 + a xM 2 ) 1 / 2 ;

[0125] where a SM is the combined acceleration of the current trajectory calculation point M.

[0126] S34. Calculate the combined acceleration of each trajectory calculation point and summarize to obtain a combined acceleration set. Select the maximum combined acceleration from the combined acceleration set. The calculation formula is

[0127] a SMAX =Max(a SM );

[0128] where a SMAX is the maximum combined acceleration among the combined accelerations corresponding to all trajectory calculation points.

[0129] S4. The trajectory equation contains multiple trajectory polynomial coefficients. Optimize and iterate each trajectory polynomial coefficient in combination with the maximum combined acceleration to obtain the optimal polynomial coefficients. Specifically, it includes the following steps.

[0130] S41. The basic parameters obtained in step S1 include the coefficient threshold of the trajectory polynomial coefficient A. Determine the initial value of the trajectory polynomial coefficient A based on the coefficient threshold. Specifically, it includes the following steps:

[0131] S411. The coefficient threshold includes the upper limit and lower limit of the trajectory polynomial coefficient A. Usually, the range of this coefficient threshold is set through empirical values.

[0132] S412. Set multiple reference points of the trajectory polynomial coefficient A within the upper limit and lower limit of the coefficient threshold.

[0133] In a specific implementation, set the range of the coefficient threshold of coefficient A as (-10 -6 , 10 -6 ), and set 13 reference points of coefficient A within the coefficient threshold range:

[0134] In this embodiment, the 13 reference points are specifically (-10 -6 , -10 -7 , -10 -8 , -10 -9 , -10 -10 , -10 -11 , 0.0, 10 -11 , 10 -10 , 10 -9 , 10 -8 , 10 -7 , 10 -6 ).

[0135] S413. Calculate the maximum comprehensive acceleration a corresponding to each reference point of coefficient A SMAX , as Figure 2 shown, "the initial values and ranges of the selected trajectory polynomial coefficients" will input the reference points of coefficient A in the previous step, and the previous step will return the calculation results, that is, the maximum comprehensive acceleration corresponding to the reference points.

[0136] S414. Screen out the minimum value from all the maximum comprehensive accelerations, and set the reference point of coefficient A corresponding to this minimum value as the initial value for the optimization calculation of the sixth-degree polynomial coefficient.

[0137] S42. Optimize and iterate the trajectory polynomial coefficients according to the preset constraint rules and initial values. The process of optimization and iteration includes

[0138] S421. Obtain the relational expressions of the trajectory polynomial coefficient B, the trajectory polynomial coefficient C, the trajectory polynomial coefficient D, the trajectory polynomial coefficient E, the trajectory polynomial coefficient F, and the trajectory polynomial coefficient G according to the constraint rules and the initial values.

[0139] The constraint rules include multiple constraint conditions

[0140] Among them, the first constraint condition is that the curve represented by the trajectory equation passes through the starting point coordinates.

[0141] In this embodiment, that is, the curve passes through the starting point O(0.0, 0.0),

[0142] From the first constraint condition, the first constraint equation is obtained:[[]]

[0143] G = 0.0;

[0144] G is the trajectory polynomial coefficient.

[0145] The second constraint condition is that the slope of the curve at the starting point coordinates is the tangent function of the initial course angle.

[0146] In this embodiment, that is, the slope of the curve at the starting point O is the tangent function of the initial course angle α0,

[0147] The second constraint condition gives the second constraint equation:

[0148] F = tg(α0);

[0149] F is the coefficient of the trajectory polynomial.

[0150] The third constraint condition is that the curve represented by the trajectory equation passes through the end point coordinates.

[0151] In this embodiment, that is, the curve passes through the end point S(Lg, Hg);

[0152] The third constraint condition gives the third constraint equation:

[0153] A×L 6 +B×L 5 +C×L 4 +D×L 3 +E×L 2 +L×tg(α0)+G = Hg;

[0154] A, B, C, D, E, F, and G are all coefficients of the trajectory polynomial.

[0155] The fourth constraint condition is that the curve slope of the end point coordinates is the tangent function of the end point course angle.

[0156] In this embodiment, that is, the curve slope of the end point S is the tangent function of the end point course angle α S ;

[0157] The fourth constraint condition gives the fourth constraint equation:

[0158] tg(α S ) = 6A×L 5 +5B×L 4 +4C×L 3 +3D×L 2 +2E×L+tg(α0).

[0159] S422. According to the first constraint condition, the second constraint condition, the third constraint condition, and the fourth constraint condition, the above four constraint equations are obtained.

[0160] Combined with the curvatures of the starting point O and the end point S, the fifth constraint equation and the sixth constraint equation are obtained in sequence.

[0161] That is, the fifth constraint equation is:

[0162] K O = 2E / (1 + F 2 ) 3 / 2 ;

[0163] Where KO is the curvature at the starting point O.

[0164] The sixth constraint equation is:

[0165] K S = (30A × L 4 + 20B × L 3 + 12C × L 2 + 6D × L + 2E) / (1 + tg 2 (α S )) 3 / 2 ;

[0166] where K S is the curvature at the end point S.

[0167] S423. Using the above six constraint equations, the following relationships for the coefficients B, C, D, E, F, and G of the trajectory polynomial can be obtained:

[0168] B = {-3A × L 6 + 6Hg - 3[tg(α0) + tg(α S )] × L + (0.5P - E) × L 2} / L 5 ;

[0169] C = {3A × L 6 - 15Hg + [8tg(α0) + 7tg(α S )] × L - (P - 3E) × L 2} / L 4 ;

[0170] D = {-A × L 6 + 10Hg - [6tg(α0) + 4tg(α S )] × L + (0.5P - 3E) × L 2} / L 3 ;

[0171] E = 0.5 × K O × [1 + tg 2 (α0)] 3 / 2 ;

[0172] F = tg(α0);

[0173] G = 0.0;

[0174] where L = Lg;

[0175] P = K S × [1 + tg 2 (α S )] 3 / 2 .

[0176] S423. Establish an optimization objective function, and obtain the optimal polynomial coefficients of the trajectory equation based on the optimization objective function. The optimal polynomial coefficients are the trajectory polynomial coefficients that minimize the maximum comprehensive acceleration. The optimization objective function is:

[0177] Y = Min(a SMAX );

[0178] where Y represents the minimum (optimal) maximum comprehensive acceleration corresponding to the optimal polynomial coefficients of the trajectory equation.

[0179] Within the coefficient threshold range of all trajectory polynomial coefficients, solve for the sixth-degree polynomial coefficients that minimize the maximum comprehensive acceleration of the trajectory.

[0180] In the above coefficient calculation process, given the coefficient A, a trajectory can be obtained. Calculate the comprehensive acceleration of each trajectory calculation point on this trajectory, and find the maximum comprehensive acceleration a SMAX , which is the maximum comprehensive acceleration of this trajectory (a SMAX ). Refer to Figure 2 , when "optimizing and calculating the sixth-degree polynomial coefficients of the trajectory", the calculation of the maximum comprehensive acceleration will be repeatedly requested.

[0181] The purpose of this optimization iteration step is to find the coefficient A that minimizes the maximum comprehensive acceleration (a SMAX ) on the corresponding trajectory.

[0182] S5. Output the optimized sixth-degree polynomial coefficients of each trajectory to the downstream planning module, update the trajectory equation in combination with the optimal polynomial coefficients, and perform trajectory planning using the updated trajectory equation, which can minimize the maximum comprehensive acceleration and thus achieve trajectory optimization.

[0183] In summary, the method of this application is a trajectory optimization algorithm that provides a trajectory optimization algorithm based on the minimization of the vehicle's comprehensive acceleration (the vector sum of the longitudinal acceleration and the lateral acceleration) for the ADAS emergency obstacle avoidance function (AES). The trajectory is expressed by a sixth-degree polynomial, and the speed planning is expressed by a third-degree polynomial. Through the constraints of the positions, speed directions, and curvatures of the starting point and the ending point, the minimization of the comprehensive acceleration is used as the optimization objective. After optimization calculation, the optimal sixth-degree polynomial coefficients of the trajectory are obtained, thereby providing a trajectory planning with the minimum comprehensive acceleration for the ADAS emergency obstacle avoidance function and improving the safety of the ADAS emergency obstacle avoidance function.

[0184] Specifically, between the starting point and the ending point of the trajectory, a certain number of trajectory calculation points are set, and the number can meet the requirements of the emergency obstacle avoidance function for calculation and control accuracy. A method for calculating the comprehensive acceleration of each trajectory calculation point is established. The comprehensive acceleration of each trajectory calculation point is the vector sum of the longitudinal acceleration and the lateral acceleration at that point. Then, the maximum comprehensive acceleration among all the calculation points from the starting point to the ending point is found. Finally, an optimization algorithm is used to find a set of coefficients of the trajectory sixth-degree polynomial to minimize the maximum comprehensive acceleration, and this set of sixth-degree polynomial coefficients is provided for emergency obstacle avoidance to perform trajectory planning.

[0185] The implementation of the method of this application is verified through the analysis and comparison of two specific cases below.

[0186] Embodiment 1:

[0187] This embodiment is the analysis of the calculation results of the sixth-degree polynomial (optimized) and the fifth-degree polynomial (non-optimized). As Figure 4 and Figure 5 shown, the following are the data of this embodiment.

[0188] Starting point O coordinates (0.0, 0.0);

[0189] Ending point S coordinates (31.0, 2.5);

[0190] Velocity planning polynomial coefficients: Q = 1.7422, R = -3.92, U = 0.0, W = 22.2222;

[0191] Velocity planning end time (Ts): 1.5;

[0192] Time to collision (TTC): 1.5;

[0193] Starting point O heading angle (α O ): 0.0;

[0194] Ending point S heading angle (α S ): 0.0;

[0195] Starting point O curvature (K O ): 0.0;

[0196] Ending point S curvature (K S ): 0.0;

[0197] Operating condition: Vehicle speed is 22.22 m / s (80 kph), and a stationary obstacle appears 31 m ahead;

[0198] Automatic emergency braking system (AEB) operating condition: Average deceleration is 0.75g, and the vehicle speed when hitting the target obstacle is 6.12 m / s (22 kph);

[0199] Automatic driving emergency steering (AES) working condition: lateral displacement of 2.5m to avoid obstacles;

[0200] Speed ​​change planning comprehensive acceleration peak value (quintic polynomial non-optimization): 7.41 (m / s²) does not meet the safety boundary;

[0201] Speed ​​change planning comprehensive acceleration peak (sixth-order polynomial optimization): 6.89 (m / s²) meets the safety margin;

[0202] The maximum peak value difference is 7.55%.

[0203] The maximum comprehensive acceleration peak after optimization is 7.55% lower than that of non-optimization.

[0204] Embodiment 2:

[0205] This example is an analysis of the calculation results of the sixth-order polynomial (optimized) and the fifth-order polynomial (non-optimized). Figure 6 and Figure 7 As shown, the data of this embodiment are as follows.

[0206] Starting point O coordinates (0.0, 0.0);

[0207] End point S coordinate (70.0, 15.0);

[0208] Speed ​​planning curve coefficients: Q=0.4444, R=-2.0, U=0.0, W=22.2222;

[0209] Speed ​​planning end time (Ts): 5.0;

[0210] Time to Collision (TTC): 3.0;

[0211] Starting point O heading angle (α O ):0.00872664626;

[0212] End point S heading angle (α S ):0.0104719755;

[0213] Starting point O curvature (K O ):0.00202479339;

[0214] End point S curvature (K S ):0.00369169101.

[0215] Working condition: The vehicle speed is 22.22m / s (80kph), a stationary obstacle appears 70m ahead, and the vehicle moves sideways 15m to avoid the obstacle.

[0216] Speed ​​change planning comprehensive acceleration peak value (quintic polynomial non-optimization): 7.19 (m / s²) does not meet the safety margin;

[0217] The peak value of the comprehensive acceleration in the variable speed planning (optimized by the sixth-degree polynomial): 6.44 (m / s²) meets the safety boundary;

[0218] The difference between the maximum peak values: 11.65%.

[0219] The peak value of the non-optimized maximum comprehensive acceleration is 11.65% larger than the optimized value.

[0220] Based on the same inventive concept described above, an embodiment of the present application also discloses a trajectory optimization device for emergency obstacle avoidance in vehicle autonomous driving, the architecture of which is as Figure 8 shown. The device includes the following modules:

[0221] The basic parameter acquisition module is configured to receive preset basic parameters, where the basic parameters include the starting point of the trajectory of the vehicle driving and the starting point parameters of the starting point, and also include the ending point of the trajectory and the ending point parameters of the ending point, and receive the pre-established speed planning equation of the trajectory;

[0222] The speed planning module is configured to establish the trajectory equation of the trajectory, set a plurality of trajectory calculation points between the starting point and the ending point according to the trajectory equation, and calculate the longitudinal speed and longitudinal acceleration of each trajectory calculation point according to the speed planning equation;

[0223] The maximum comprehensive acceleration calculation module is configured to calculate the comprehensive acceleration of each trajectory calculation point and obtain a comprehensive acceleration set by combining the trajectory equation and the speed and longitudinal acceleration of each trajectory calculation point, and screen the maximum comprehensive acceleration from the comprehensive acceleration set;

[0224] The coefficient optimization module is configured to the trajectory equation includes a plurality of trajectory polynomial coefficients, optimize and iterate each trajectory polynomial coefficient by combining the maximum comprehensive acceleration to obtain the optimal polynomial coefficients, update the trajectory equation by combining the optimal polynomial coefficients, and the updated trajectory equation is used for trajectory planning.

[0225] In a specific feasible implementation, the speed planning module includes the following units:

[0226] The first speed planning unit is configured to establish the trajectory equation of the trajectory, where the trajectory is the driving curve of the vehicle from the starting point to the ending point, the trajectory equation is a multi-degree polynomial equation, and the trajectory equation includes a plurality of trajectory polynomial coefficients.

[0227] The second speed planning unit is configured to set a plurality of trajectory calculation points between the starting point and the ending point according to the trajectory equation, all the trajectory calculation points are located on the trajectory, and calculate the first derivative and the second derivative of each trajectory calculation point respectively;

[0228] The trajectory equation is,

[0229] Y = A × XM 6 +B×X M 5 +C×X M 4 +D×X M 3 +E×X M 2 +F×X M +G,

[0230] wherein, X M is the longitudinal coordinate of any trajectory calculation point on the trajectory between the starting point and the end point, Y is the transverse coordinate of the trajectory calculation point corresponding to X M and A, B, C, D, E, F, G are the trajectory polynomial coefficients of the trajectory equation;

[0231] The calculation formulas for the first derivative and the second derivative of the trajectory calculation point are

[0232] Y’ = 6A×X M 5 + 5B×X M 4 + 4C×X M 3 + 3D×X M 2 + 2E×X M + F;

[0233] Y’’ = 30A×X M 4 + 20B×X M 3 + 12C×X M 2 + 6D×X M + 2E;

[0234] wherein, Y’ is the first derivative of the trajectory calculation point corresponding to X M and Y’’ is the second derivative of the trajectory calculation point corresponding to X M .

[0235] The third velocity planning unit is used to calculate the longitudinal velocity and longitudinal acceleration of each trajectory calculation point in the trajectory according to the velocity planning equation. The velocity planning equation includes

[0236] V M = Q×t 3 + R×t 2 + U×t + W;

[0237] a xM = 3×Q×t 2 + 2×R×t + U;

[0238] where \(t\) is any moment after the vehicle starts obstacle avoidance, \(Q\), \(R\), \(U\), and \(W\) are the coefficients of the velocity planning polynomial, and \(V\) M is the longitudinal velocity of the trajectory calculation point corresponding to the moment \(t\), and \(a\) xM is the longitudinal acceleration of the trajectory calculation point corresponding to the moment \(t\).

[0239] In a specific feasible implementation, the maximum comprehensive acceleration calculation module includes the following units:

[0240] The first maximum comprehensive acceleration calculation unit is used to calculate the curvature \(K\) of the current trajectory calculation point based on the first derivative and the second derivative of any trajectory calculation point M ;

[0241] The second maximum comprehensive acceleration calculation unit is used to calculate the lateral acceleration \(a\) of the current trajectory calculation point based on the longitudinal velocity and the curvature of the current trajectory calculation point yM , and the calculation formula is

[0242] a yM = V M 2 ×K M ;

[0243] The third maximum comprehensive acceleration calculation unit is used to calculate the comprehensive acceleration \(a\) of the current trajectory calculation point based on the lateral acceleration and the longitudinal acceleration of the current trajectory calculation point SM , and the calculation formula is

[0244] a SM = (a yM 2 + a xM 2 ) 1 / 2 ;

[0245] The fourth maximum comprehensive acceleration calculation unit is used to calculate the comprehensive acceleration of each trajectory calculation point and aggregate them to obtain a comprehensive acceleration set, and screen the maximum comprehensive acceleration from the comprehensive acceleration set. The calculation formula is

[0246] a SMAX = Max(a SM );

[0247] where \(a\) SMAX is the maximum comprehensive acceleration among the comprehensive accelerations corresponding to all trajectory calculation points.

[0248] In a specific feasible implementation, the coefficient optimization module includes the following units:

[0249] The first coefficient optimization unit is used for the basic parameters to further include a coefficient threshold of the trajectory polynomial coefficient A, and determines the initial value of the trajectory polynomial coefficient A according to the coefficient threshold.

[0250] The second coefficient optimization unit is used to optimize and iterate each trajectory polynomial coefficient according to the preset constraint rules and the initial value. The process of optimization iteration includes:

[0251] Obtain the relational expressions of the trajectory polynomial coefficients B, C, D, E, F, and G according to the constraint rules and the initial value, establish the optimization objective function, and obtain the optimal polynomial coefficients of the trajectory equation according to the optimization objective function.

[0252] In a specific feasible implementation, the second coefficient optimization unit includes the following sub-units:

[0253] The coefficient optimization sub-unit is used for the starting point parameters to include the starting point coordinates, starting point curvature, and initial heading angle, and the ending point parameters to include the ending point coordinates, ending point curvature, and ending point heading angle.

[0254] The constraint rules include multiple constraint conditions.

[0255] Among them, the first constraint condition is that the curve represented by the trajectory equation passes through the starting point coordinates.

[0256] The second constraint condition is that the curve slope of the starting point coordinates is the tangent function of the initial heading angle.

[0257] The third constraint condition is that the curve represented by the trajectory equation passes through the ending point coordinates.

[0258] The fourth constraint condition is that the curve slope of the ending point coordinates is the tangent function of the ending point heading angle.

[0259] According to the first constraint condition, the second constraint condition, the third constraint condition, and the fourth constraint condition, combined with the curvatures of the starting point and the ending point, obtain the relational expressions of the trajectory polynomial coefficients B, C, D, E, F, and G.

[0260] From the above function introduction, it can be seen that a trajectory optimization device for emergency obstacle avoidance in vehicle autonomous driving in this application builds a complete trajectory optimization system for emergency obstacle avoidance, improves the intelligent and automated levels of vehicle autonomous driving emergency obstacle avoidance technology, enhances the safety and stability of the obstacle avoidance process, and promotes the sustainable, rapid, and healthy development of the economic society.

[0261] Based on the same inventive concept described above, an embodiment of the present application also discloses a computer-readable storage medium, in which at least one instruction, at least one program, a code set or an instruction set is stored, and the at least one instruction, at least one program, the code set or the instruction set can be loaded and executed by a processor to implement the trajectory optimization method for emergency obstacle avoidance in automotive autonomous driving provided in the above method embodiment.

[0262] Similarly based on the same inventive concept described above, an embodiment of the present application also discloses a computer-readable storage medium, in which at least one instruction, at least one program, a code set or an instruction set is stored, and the at least one instruction, at least one program, the code set or the instruction set is loaded and executed by a processor to implement the trajectory optimization method for emergency obstacle avoidance in automotive autonomous driving as described above.

[0263] Those of ordinary skill in the art can understand that all or part of the steps of implementing the above embodiments can be completed by hardware, or can be completed by instructing relevant hardware through a program. The program can be stored in the computer-readable storage medium, and the computer-readable storage medium includes, for example: various media that can store program codes such as USB flash drives, mobile hard disks, read-only memories (ROMs), random access memories (RAMs), magnetic disks or optical discs.

[0264] The above are only the optional embodiments of the present application, and are not intended to limit the present application. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A trajectory optimization method for emergency obstacle avoidance in automotive autonomous driving, characterized in that, It includes the following steps: Receive preset basic parameters, where the basic parameters include the starting point of the trajectory of the vehicle's travel and the starting point parameters of the starting point, and also include the ending point of the trajectory and the ending point parameters of the ending point, and receive the pre-established speed planning equation of the trajectory; Establish the trajectory equation of the trajectory, set multiple trajectory calculation points between the starting point and the ending point according to the trajectory equation, and calculate the longitudinal speed and longitudinal acceleration of each trajectory calculation point according to the speed planning equation; Calculate the comprehensive acceleration of each trajectory calculation point and obtain a comprehensive acceleration set by combining the trajectory equation and the longitudinal speed and longitudinal acceleration of each trajectory calculation point, and screen the maximum comprehensive acceleration from the comprehensive acceleration set. Specifically, it includes the following steps Obtain the curvature of the current trajectory calculation point according to any trajectory calculation point, and calculate the lateral acceleration of the current trajectory calculation point according to the longitudinal speed and the curvature of the current trajectory calculation point; The trajectory equation contains multiple trajectory polynomial coefficients. Optimize and iterate each trajectory polynomial coefficient by combining the maximum comprehensive acceleration to obtain the optimal polynomial coefficients, and update the trajectory equation by combining the optimal polynomial coefficients. The updated trajectory equation is used for trajectory planning.

2. The trajectory optimization method for emergency obstacle avoidance in autonomous driving of an automobile according to claim 1, characterized in that The steps of establishing the trajectory equation of the trajectory and setting multiple trajectory calculation points between the starting point and the ending point according to the trajectory equation specifically include the following steps: Establish the trajectory equation of the trajectory. The trajectory is the driving curve of the vehicle from the starting point to the ending point. The trajectory equation is a polynomial equation of multiple degrees, and the trajectory equation contains multiple trajectory polynomial coefficients; Set multiple trajectory calculation points between the starting point and the ending point according to the trajectory equation. All the trajectory calculation points are located on the trajectory, and calculate the first derivative and the second derivative of each trajectory calculation point respectively.

3. The trajectory optimization method for emergency obstacle avoidance in vehicle autonomous driving according to claim 2, wherein: The trajectory equation is Y = A×X M 6 + B×X M 5 + C×X M 4 + D×X M 3 + E×X M 2 + F×X M + G Among them, X M is the longitudinal coordinate of any trajectory calculation point on the trajectory between the starting point and the ending point, and Y is the transverse coordinate of the trajectory calculation point corresponding to X M and A, B, C, D, E, F, G are the trajectory polynomial coefficients of the trajectory equation; The calculation formulas for the first derivative and the second derivative of the trajectory calculation point are Y’ = 6A × X M 5 + 5B × X M 4 + 4C × X M 3 + 3D × X M 2 + 2E × X M + F; Y’’ = 30A×X M 4 + 20B×X M 3 + 12C×X M 2 + 6D×X M + 2E; where Y’ is the first derivative of the trajectory calculation point corresponding to X M and Y’’ is the second derivative of the trajectory calculation point corresponding to X M ​ 4. The trajectory optimization method for emergency obstacle avoidance in autonomous driving of an automobile according to claim 3, characterized in that, The steps of calculating the longitudinal speed and longitudinal acceleration of each trajectory calculation point according to the speed planning equation specifically include the following steps: The speed planning equation includes V M = Q × t 3 + R × t 2 + U × t + W; a xM = 3×Q×t 2 + 2×R×t + U; Among them, t is any moment after the vehicle starts obstacle avoidance, Q, R, U, and W are the coefficients of the speed planning polynomial, and V M is the longitudinal speed of the trajectory calculation point corresponding to the moment t, and a xM is the longitudinal acceleration of the trajectory calculation point corresponding to the moment t; Calculate the longitudinal speed and longitudinal acceleration of each trajectory calculation point in the trajectory according to the speed planning equation.

5. The trajectory optimization method for emergency obstacle avoidance in autonomous driving of an automobile according to claim 4, wherein The steps of calculating the comprehensive acceleration of each trajectory calculation point and obtaining a comprehensive acceleration set by combining the trajectory equation and the speed and longitudinal acceleration of each trajectory calculation point, and screening the maximum comprehensive acceleration from the comprehensive acceleration set also include the following steps: Calculate the first derivative and the second derivative of the point according to any of the described trajectories, and obtain the curvature K of the currently described trajectory calculation point M ; The lateral acceleration a of the currently described trajectory calculation point yM The calculation formula is a yM =V M 2 ×K M ; Calculate the lateral acceleration and longitudinal acceleration of the point according to the current trajectory, and calculate the comprehensive acceleration a of the current trajectory calculation point SM , and the calculation formula is a SM =(a yM 2 + a xM 2 ) 1 / 2 ; Calculate the comprehensive acceleration of each trajectory calculation point and summarize to obtain a comprehensive acceleration set, and screen the maximum comprehensive acceleration from the comprehensive acceleration set. The calculation formula is a SMAX =Max(a SM ); where a SMAX is the maximum comprehensive acceleration among the comprehensive accelerations corresponding to all the trajectory calculation points.

6. The trajectory optimization method for emergency obstacle avoidance in autonomous driving of an automobile according to claim 3, characterized in that, The steps of optimizing and iterating each trajectory polynomial coefficient by combining the maximum comprehensive acceleration to obtain the optimal polynomial coefficients specifically include the following steps: The basic parameters further include a coefficient threshold of the coefficient A of the trajectory polynomial, and an initial value of the coefficient A of the trajectory polynomial is determined according to the coefficient threshold; Each of the trajectory polynomial coefficients is optimized and iterated according to a preset constraint rule and the initial value. The process of the optimization iteration includes obtaining relationships of the trajectory polynomial coefficient B, the trajectory polynomial coefficient C, the trajectory polynomial coefficient D, the trajectory polynomial coefficient E, the trajectory polynomial coefficient F, and the trajectory polynomial coefficient G according to the constraint rule and the initial value, establishing an optimization objective function, and obtaining optimal polynomial coefficients of the trajectory equation according to the optimization objective function.

7. The trajectory optimization method for emergency obstacle avoidance in vehicle autonomous driving according to claim 6, characterized in that: The starting point parameters include the starting point coordinates, the starting point curvature, and the initial heading angle, and the ending point parameters include the ending point coordinates, the ending point curvature, and the ending point heading angle; The constraint rule includes a plurality of constraint conditions, wherein, the first constraint condition is that the curve represented by the trajectory equation passes through the starting point coordinates; The second constraint condition is that the curve slope of the starting point coordinates is the tangent function of the initial heading angle; The third constraint condition is that the curve represented by the trajectory equation passes through the ending point coordinates; The fourth constraint condition is that the curve slope of the ending point coordinates is the tangent function of the ending point heading angle; According to the first constraint condition, the second constraint condition, the third constraint condition, and the fourth constraint condition, and combining the curvatures of the starting point and the ending point, relationships of the trajectory polynomial coefficients B, C, D, E, F, and G are obtained.

8. A trajectory optimization device for emergency obstacle avoidance in autonomous driving of an automobile, characterized in that It includes the following modules: A basic parameter acquisition module, configured to receive preset basic parameters, where the basic parameters include the starting point of the trajectory of the vehicle driving and the starting point parameters of the starting point, and further include the ending point of the trajectory and the ending point parameters of the ending point, and receive the pre-established speed planning equation of the trajectory; A speed planning module, configured to establish a trajectory equation of the trajectory, set a plurality of trajectory calculation points between the starting point and the ending point according to the trajectory equation, and calculate the longitudinal speed and longitudinal acceleration of each of the trajectory calculation points according to the speed planning equation; The maximum comprehensive acceleration calculation module is used to calculate the comprehensive acceleration of each trajectory calculation point by combining the trajectory equation and the speed and longitudinal acceleration of each trajectory calculation point, and obtain a comprehensive acceleration set, and screen the maximum comprehensive acceleration from the comprehensive acceleration set; specifically, according to any trajectory calculation point, the curvature K of the current trajectory calculation point is obtained M , and according to the longitudinal speed and the curvature of the current trajectory calculation point, the lateral acceleration a of the current trajectory calculation point is calculated yM ; A coefficient optimization module, where the trajectory equation includes a plurality of trajectory polynomial coefficients, and each of the trajectory polynomial coefficients is optimized and iterated in combination with the maximum comprehensive acceleration to obtain optimal polynomial coefficients, and the trajectory equation is updated in combination with the optimal polynomial coefficients, and the updated trajectory equation is used for trajectory planning.

9. An intelligent terminal, characterized in that It includes a memory and a processor. At least one program is stored in the memory, and the at least one program is loaded and executed by the processor to implement the trajectory optimization method for emergency obstacle avoidance in vehicle autonomous driving according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, At least one program is stored in the readable storage medium, and the at least one program is loaded and executed by the processor to implement the trajectory optimization method for emergency obstacle avoidance in vehicle autonomous driving according to any one of claims 1 to 7.

Images