An intelligent automobile trajectory planning method based on an indicator

Abstract

This invention provides a trajectory planning method for intelligent vehicles based on indicator signs, comprising path planning and speed planning processes: Path planning includes establishing hypothetical bypass point coordinates based on basic data information containing pile location information and safe distance information, forming a hypothetical path based on the hypothetical bypass point coordinates, performing collision optimization on the hypothetical path to obtain preferred bypass points, and planning a target path passing through the preferred bypass points; Speed ​​planning includes calculating the coordinates of a pre-aiming point measured with a reference pre-aiming time based on the path planning and vehicle operation information, and correcting the vehicle's operating speed based on the pre-aiming point coordinates. This invention's planning method can solve the planning problem for intelligent driving vehicles in environments with indicator signs guiding vehicle movement, thus broadening the application scenarios of intelligent driving vehicles.

Description

Technical Field

[0001] This invention belongs to the field of intelligent vehicle path planning and design technology, specifically relating to an intelligent vehicle trajectory planning method based on indicator signs. Background Technology

[0002] In recent years, with the rapid development of sensor technology and the continuous breakthroughs in artificial intelligence, intelligent driving has gradually become an extremely important development direction in the automotive field, and has made substantial progress. The continuous updating and improvement of core algorithms such as perception, planning, and control have enabled intelligent driving vehicles to handle an increasing number of scenarios, and to handle them more intelligently. Commonly used perception, planning, and control algorithms in the industry can handle common situations involving vehicles, pedestrians, bicycles, static obstacles, and traffic signs, covering the vast majority of users' daily usage scenarios.

[0003] However, given the complex and ever-changing traffic and usage environments, some special scenarios still require special handling. Examples include test sites for placing indicator cones, road construction sites for placing indicator cones, and industrial parks / ports where indicator signs / cone cones are placed. The problem this invention aims to solve is the motion planning problem for intelligent driving vehicles in environments with guiding vehicle signs.

[0004] Currently, the existing patents related to this invention are as follows: Patent application CN 111942387A obtains road information and information about vehicles ahead of the vehicle's current route, then plans the vehicle's lane-changing route based on this information, and outputs a signal to display the lane-changing route in association with the road information; Invention patent CN101868397 B provides a path calculation based on the geometric conditions of the parking space and the vehicle's position relative to the parking space, including first and second path curves; CN 111857118 A generates painter descriptor data describing the environment surrounding the autonomous vehicle, and based on this data generates a parking trajectory for the autonomous vehicle consisting of one or more trajectory segments to avoid collisions with obstacles; Invention CN 111169469A fits trajectory equations based on lateral deviation, lateral velocity, and lateral acceleration to obtain a trajectory curve, and then performs autonomous driving based on the trajectory curve. The above patents can provide trajectory planning based on common vehicle operating environments. However, they cannot solve the planning problem for intelligent driving vehicles in environments with signs guiding vehicle movement.

[0005] Invention Patent Content

[0006] To address the aforementioned problems, this invention provides a trajectory planning method for intelligent driving vehicles based on directional signs, aiming to solve the trajectory planning problem of intelligent driving vehicles in environments with directional signs guiding vehicle movement. This objective is achieved through the following technical solution:

[0007] This invention provides an intelligent vehicle trajectory planning method based on indicator markers, which includes path planning and speed planning processes: Path planning includes establishing assumed bypass point coordinates based on basic data information of pile points containing pile point location information and pile point safety distance information, forming an assumed path based on the assumed bypass point coordinates, performing collision optimization on the assumed path to obtain preferred bypass points, and planning a target path passing through the preferred bypass points; Speed ​​planning includes calculating the coordinates of the aiming point measured by the reference aiming time based on the path planning and vehicle operation information, and correcting the vehicle operating speed based on the aiming point coordinates.

[0008] Preferably, the path planning includes the following steps:

[0009] Pile point description - Identify pile points and obtain pile point location information, pile point size, and the minimum safe distance that vehicles should maintain between the pile points when detouring around them;

[0010] Description of the detour point - The detour direction is determined based on the location information of the detour point, the detour distance is set according to the minimum safe distance for detour around the detour point, and the hypothetical detour point is established based on the detour point data information including the detour direction and detour distance. The detour distance refers to the distance between the vehicle and the detour point when detouring the detour point.

[0011] Path Description - The path between adjacent assumed stake points is set as the assumed path. The assumed path is decomposed into multiple basic paths. A horizontal line graph is plotted using the path length and x-coordinate of each basic path segment. A vertical line graph is plotted using the path length and y-coordinate of the starting point of each basic path segment. Spline interpolation is performed on the horizontal and vertical line graphs to obtain the horizontal and vertical curves respectively. The horizontal and vertical curves are then merged to form the planned path.

[0012] Collision optimization - Take multiple collision feature points on the planned path, calculate the expected distance between each collision feature point and the pile point. When the expected distance of any collision feature point is less than the minimum safe distance for the pile point to bypass, return to the pile point description step, reset the assumed pile point with the corrected pile distance and / or pile angle, until all collision feature points on the planned path meet the requirements, and generate the target path.

[0013] Preferably, the path planning further includes a curvature optimization step, which is set between the path description step and the collision optimization step, specifically:

[0014] Curvature optimization - On the planned path, take multiple curvature feature points and calculate their curvature radii respectively. When the curvature radius of any curvature feature point is less than the threshold value of the preset driving radius when the vehicle performs a turning operation, jump to the slalom point description step, reset the assumed slalom point with the corrected slalom distance and / or slalom angle, and perform path description and curvature optimization until the curvature radius of each curvature feature point is greater than the threshold value. The initial planned path that meets the requirements is set as the preferred planned path.

[0015] Collision optimization - On the preferred planning path obtained in the curvature optimization step, take multiple collision feature points, calculate the expected distance between each collision feature point and the pile point. When the expected distance of any collision feature point is less than the minimum safe distance for the pile point to bypass, return to the pile point description step, reset the assumed pile point with the corrected pile distance and / or pile angle, and then perform the curvature optimization and collision optimization steps until all curvature feature points and collision feature points on the preferred planning path meet the requirements, and generate the target path.

[0016] Furthermore, the velocity planning process includes the following steps when calculating acceleration: solving for the aiming point, solving for alternative accelerations, and dynamic verification.

[0017] Pre-aiming point solution - Set the coordinates of the next pre-aiming point along the target path, using the baseline pre-aiming time and current speed as references;

[0018] Alternative acceleration solution - Based on multiple alternative aiming times with reference to the baseline aiming time, calculate multiple alternative accelerations to reach the next aiming point. The alternative accelerations include lateral acceleration values ​​and longitudinal acceleration values.

[0019] Dynamics verification - Multiple candidate accelerations are generated by corresponding the lateral and longitudinal acceleration values ​​to the corresponding positions of the x and y coordinates of the GG graph and filled into the GG graph. The preferred acceleration points that fall into the preset dynamics region of the GG graph are then selected.

[0020] If multiple preferred acceleration points are obtained from the dynamics verification, the target acceleration is obtained by selecting one according to the preset point selection rules.

[0021] Furthermore, the preset point selection rules include steps of speed evaluation and speed selection, specifically:

[0022] Speed ​​evaluation - The assignment of values ​​to the candidate acceleration corresponding to the preferred acceleration point falling into the preset dynamic region in terms of vehicle driving comfort, safety and efficiency indicators;

[0023] Speed ​​selection—Based on the preset weights of comfort, safety and efficiency indicators, the acceleration equilibrium point with the best balance performance is measured by weighted average method. The acceleration corresponding to the acceleration equilibrium point is defined as the target acceleration, and the expected operating speed of the vehicle is measured based on the target acceleration.

[0024] Furthermore, in the pile point description step, a reference line for the pile angle extending radially through the pile point is set using the included angle between the pile points as a parameter. The reference line for the pile angle is deflected by a preset pile angle and intersects with a circle with the pile distance as the radius and the pile point as the center to generate a hypothetical pile point.

[0025] Furthermore, the pre-aiming point solution includes the steps of calculating the pre-aiming distance, setting the pre-aiming point coordinates based on the pre-aiming distance, and converting the pre-aiming point coordinates into the horizontal coordinates and vertical coordinates of the pre-aiming point with the vehicle as the origin.

[0026] The beneficial effects of this invention are as follows:

[0027] (1) The planning method of the present invention can solve the planning problem of intelligent driving vehicles in an environment where there are signs to guide the vehicles to drive, and broaden the application scenarios of intelligent driving vehicles. These specific application environments include: test sites where signs are placed, road construction sites where signs are placed, parks / ports where signs / signs are placed, etc.

[0028] (2) It can perform trajectory planning using a small number of piles / markers and corresponding pile direction marks, thereby improving the intelligence level of the intelligent driving vehicle planning algorithm. Attached Figure Description

[0029] Figure 1 The present invention provides a flowchart of an intelligent driving vehicle trajectory planning method based on indicator tags.

[0030] Figure 2 This is a schematic diagram illustrating the detour points of the intelligent driving vehicle trajectory planning method based on indicator markers proposed in this invention.

[0031] Figure 3 This is a schematic diagram illustrating the path description of the intelligent driving vehicle trajectory planning method based on indicator tags proposed in this invention.

[0032] Figure 4 This is a schematic diagram illustrating the curvature optimization of the intelligent driving vehicle trajectory planning method based on indicator markers proposed in this invention.

[0033] Figure 5 This is a schematic diagram illustrating the collision optimization of the intelligent driving vehicle trajectory planning method based on indicator markers proposed in this invention.

[0034] Figure 6This is a schematic diagram illustrating the pre-aiming point solution for the intelligent driving vehicle trajectory planning method based on indicator markers proposed in this invention;

[0035] Figure 7 A schematic diagram illustrating the dynamic verification of the intelligent driving vehicle trajectory planning method based on indicator markers proposed in this invention;

[0036] Figure 8 This is a schematic diagram illustrating the pile position and detour requirements of an embodiment of the intelligent driving vehicle trajectory planning method based on indicator signs proposed in this invention.

[0037] Figure 9 This is a path planning diagram of an embodiment of the intelligent driving vehicle trajectory planning method based on indicator tags proposed in this invention;

[0038] Figure 10 This is a speed planning diagram of an embodiment of the intelligent driving vehicle trajectory planning method based on indicator tags proposed in this invention. Detailed Implementation

[0039] The preferred embodiments of the present invention will now be described with reference to the accompanying drawings.

[0040] It should be noted that in the following description, the terms "upper," "lower," "top," "bottom," "inner," and "outer," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of the present invention. Furthermore, in the description of the present invention, "a plurality of" means at least two, such as two, three, etc., unless otherwise explicitly specified.

[0041] Example 1:

[0042] This invention provides an intelligent vehicle trajectory planning method based on indicator markers, which includes path planning and speed planning processes: Path planning includes establishing assumed bypass point coordinates based on basic data information of pile points containing pile point location information and pile point safety distance information, forming an assumed path based on the assumed bypass point coordinates, performing collision optimization on the assumed path to obtain preferred bypass points, and planning a target path passing through the preferred bypass points; Speed ​​planning includes calculating the coordinates of the aiming point measured by the reference aiming time based on the path planning and vehicle operation information, and correcting the vehicle operating speed based on the aiming point coordinates.

[0043] Preferably, the path planning and speed planning each include the following steps, such as... Figure 1 As shown:

[0044] Pile point description - Identify pile points and obtain pile point location information, pile point size, and minimum safe distance for pile point detour;

[0045] Description of the pile-around point - The direction of the pile-around is determined based on the location information of the pile point, and the distance of the pile-around is set according to the minimum safe distance of the pile-around. The assumed pile-around point is established based on the pile-around point data information including the direction of the pile-around and the distance of the pile-around. Preferably, the reference line of the pile-around angle extending radially through the pile point is set with the included angle between the pile points as a parameter, and the reference line of the pile-around angle is deflected by a preset pile-around angle and intersected with a circle with the distance of the pile-around as the radius and the pile point as the center to generate the assumed pile-around point.

[0046] Path Description - The path between adjacent assumed stake points is set as the assumed path. The assumed path is decomposed into multiple basic paths. A horizontal line graph is plotted using the path length of each basic path and the x-coordinate of the starting point of that basic path. A vertical line graph is plotted using the path length of each basic path and the y-coordinate of the starting point of that basic path. Spline interpolation is performed on the horizontal and vertical line graphs to obtain the horizontal and vertical curves respectively. The horizontal and vertical curves are then merged to form the planned path.

[0047] Curvature optimization - On the planned path, take multiple curvature feature points and calculate their curvature radii respectively. When the curvature radius of any curvature feature point is less than the threshold value of the preset driving radius, jump to the cone deflection point description step, reset the assumed cone deflection point with the corrected cone deflection distance and / or cone deflection angle, and perform path description and curvature optimization until the curvature radius of each curvature feature point is greater than the threshold value. The initial planned path that meets the requirements is set as the preferred planned path.

[0048] Collision optimization - On the preferred planning path, take collision feature points and calculate the expected distance between each collision feature point and the pile point. When the expected distance is less than the minimum safe distance for the pile point to bypass, return to the pile point description step, reset the assumed pile point with the corrected pile distance and / or pile angle, and then perform curvature optimization and collision optimization steps until all curvature feature points and collision feature points on the preferred planning path meet the requirements, and generate the target path.

[0049] Pre-aiming point solution - along the target path, set the coordinates of the next pre-aiming point based on the reference pre-aiming time and the current speed; preferably, this step includes calculating the pre-aiming distance, setting the pre-aiming point coordinates based on the pre-aiming distance, and converting the pre-aiming point coordinates into the abscissa and ordinate of the pre-aiming point with the vehicle as the origin;

[0050] Alternative acceleration solution - along the target path, based on multiple alternative pre-aiming times with reference to the baseline pre-aiming time, calculate multiple alternative accelerations to reach the next pre-aiming point. The alternative accelerations include lateral acceleration values ​​and longitudinal acceleration values.

[0051] Dynamics verification - Multiple candidate accelerations are generated by corresponding the lateral and longitudinal acceleration values ​​to the corresponding positions of the x and y coordinates of the GG graph and filled into the GG graph. The preferred acceleration points that fall into the preset dynamics region of the GG graph are then selected.

[0052] If multiple preferred acceleration points are obtained from the dynamics verification, the target acceleration is obtained by selecting one according to the preset point selection rules.

[0053] Speed ​​evaluation - assigning values ​​to the candidate acceleration corresponding to the preferred acceleration point falling into the preset dynamic region in terms of vehicle driving comfort, safety and efficiency indicators;

[0054] Speed ​​selection—Based on the preset weights of comfort, safety and efficiency indicators, the acceleration equilibrium point with the best balance performance is measured by weighted average method. The acceleration corresponding to the acceleration equilibrium point is defined as the target acceleration, and the expected operating speed of the vehicle is measured based on the target acceleration.

[0055] Example 2:

[0056] This embodiment 2 provides a trajectory planning method for simultaneously identifying multiple stake points. Similar to embodiment 1, the method in this embodiment mainly consists of two steps: path planning and velocity planning. The trajectory planning process is as follows:

[0057] 1. Path planning

[0058] Path planning is used to plan a route around the cones based on their locations. The path planning mainly consists of the following steps:

[0059] 1.1 Pile Point Description

[0060] The aforementioned stake points are devices used in the automotive industry to guide vehicle movement and provide guidance to the driver. One specific embodiment of a stake point is a stake placed during a serpentine test for operational stability. Another specific embodiment of a stake point is a stake or indicator used to guide vehicles to avoid obstacles during road driving.

[0061] The description of the pile points refers to their location and size, safety distance, and direction of travel around the piles. See [link / reference needed]. Figure 2 As shown, the i-th decoration point can be represented by the data structure (x). i ,y i ,r imin direct i ) describes, where x i ,y i Let r be the x-axis and y-axis coordinates of the i-th pile point in the geodetic coordinate system; iminThis represents the minimum safe distance from the i-th cone, which is the minimum distance the vehicle needs to maintain when driving around the cones; direct i This is a marker indicating the direction of the vehicle's movement around the cones, used to indicate whether the vehicle should move around the cones from the left or right side of the i-th cone. i = 1, 2…N, where i is the cone number and N is the number of cones.

[0062] 1.2 Description of the pile points

[0063] The point around the cone is a point that the vehicle passes through while driving around the cone.

[0064] See Figure 2 As shown, the description of the bypass point of the i-th pile is a description of the location and control parameters of the bypass point; the i-th bypass point can be represented by the data structure (xr i ,yr i ,r i ,θ i Description. Where xr i ,yr i Let r be the x-axis and y-axis coordinates of the i-th circumferential point in the geodetic coordinate system; i θ represents the distance maintained between the vehicle and the i-th marker when driving around it. i This is the angle between the line connecting the two pile points and the reference line for the pile angle. (xr) i ,yr i ,r i ,θ i The determination method for ) is as follows:

[0065] a) Using the coordinates of pile point i, B(x) i ,y i Using ) as the center and the preset value r i Draw a circle with radius r; the preset value r i The minimum safe distance r between pile point i and pile point i is not less than imin ;

[0066] b) Connect the coordinates of pile point i-1, A(x) i-1 ,y i-1 ) and the coordinates of pile point i B(x) i ,y i Obtain the straight line AB; connect the coordinates of the stake point i to B(x). i ,y i ) and the coordinates of pile point i+1 C(x) i+1 ,y i+1 We obtain the straight line BC; we take the angle bisector of ∠ABC as the reference line for the angle around the stake;

[0067] c) Draw a ray BD, wherein the ray satisfies: i) the starting point of the ray is point B; ii) the angle between the ray and the reference line for the pile angle is a preset value θ.i iii) The ray is located to the upper left, upper right, lower left, or lower right of the reference line for the angle around the pile, indicated by the direction marker "direct". i Determined; for example, in one specific embodiment, the direction of travel around the stake is identified as direct. i To detour around the pile to the left of pile point i, then as follows: Figure 2 As shown, ray BD is located to the upper left of the reference line for the pile angle. Ray BD intersects the circle drawn in step a) at point E(xr). i ,yr i If the coordinates of the point around the pile corresponding to pile point i are (xr), then the coordinates of the point around the pile are (xr). i ,yr i )

[0068] 1.3 Path Description

[0069] See Figure 3 As shown, the path description is defined by two curves, sx and sy, which describe the path around the stake, where s is the path length and x and y are the coordinates of the stake points in the geodetic coordinate system. The specific description method is as follows:

[0070] a) Based on the coordinates (xr) of the staking point i i ,yr i Find the path length sr corresponding to the point around the stake. i The calculation method is as follows:

[0071] sr i =sr i-1 +((xr i -xr i-1 ) 2 +(yr i -yr i-1 ) 2 ) 1 / 2 ;

[0072] in,

[0073] sr i-1 This is the path length corresponding to the previous stake point;

[0074] xr i-1 yr i-1 The coordinates of point i-1 around the pile in the geodetic coordinate system;

[0075] b) See Figure 3 As shown in b, with (sr i ,xr i Connecting the lines yields the broken line L0_s-x, which describes the relationship between the path s and x around the stake point; see reference. Figure 3 As shown in c, with (sr i ,yr iConnecting the lines yields the broken line L0_s-y, which describes the relationship between the path s and y around the stake point;

[0076] c) See Figure 3 d and Figure 3 As shown in e, spline interpolation is performed on L0_s-x and L0_s-y respectively to obtain two smooth curves L1_s-x and L1_s-y; the spline interpolation is a commonly used interpolation method in the field of mathematics.

[0077] d) Based on the x and y values ​​in the L1_s-x and L1_s-y curves, recalculate the path length as follows:

[0078] s k =s k-1 +((x k -x k-1 ) 2 +(y k -y k-1 ) 2 ) 1 / 2

[0079] in,

[0080] k = 1, 2, ..., M, where k is the index of the interpolation point during spline interpolation, and M is the number of interpolation points.

[0081] x k-1 Let x be the x-coordinate of the (k-1)th point in the L1_s-x curve in the geodetic coordinate system.

[0082] y k-1 Let y be the y-coordinate of the (k-1)th point in the L1_s-y curve in the geodetic coordinate system.

[0083] x k Let x be the x-coordinate of the k-th point in the L1_s-x curve in the geodetic coordinate system.

[0084] y k Let y be the y-coordinate of the k-th point in the L1_s-y curve in the large coordinate system.

[0085] s k-1 Let L1_s-x and L1_s-y be the path length of the (k-1)th point in the curves L1_s-x and L1_s-y.

[0086] s k Let be the path length of the k-th point in the L1_s-x and L1_s-y curves;

[0087] e) Finally, we get (x) k ,y k ,s k (See also) Figure 3 As shown in f, with s k x is the x-coordinatek The curve L2_s-x is obtained by using the ordinate; see [reference]. Figure 3 As shown in g, with s k y is the x-axis. k Using the ordinate as the vertical axis, we obtain the curve L2_s-y;

[0088] 1.4 Curvature Optimization

[0089] According to the path description (x) k ,y k ,s k Calculate the radius of curvature R at each point along the path around the pile near pile point i. k ;

[0090] x k '=(x k+1 -x k ) / (s k+1 -s k );

[0091] x k =(x k+1 '-x k ') / (s k+1 -s k );

[0092] y k '=(y k+1 -y k ) / (s k+1 -s k );

[0093] y k = (y k+1 '-y k ') / (s k+1 -s k );

[0094] R k =((x) k ' 2 +y k ' 2 ) 1.5 ) / (x k '*y k "-x k *y k ');

[0095] in,

[0096] x k+1 Let x be the x-coordinate of the (k+1)th point in the L1_s-x curve in the geodetic coordinate system.

[0097] y k+1Let y be the y-coordinate of the (k+1)th point in the L1_s-y curve in the geodetic coordinate system.

[0098] s k+1 Let L1_s-x and L1_s-y be the path lengths of the (k+1)th points in the curves L1_s-x and L1_s-y.

[0099] When R k小 At the preset threshold value R min If the radius is too small, the curvature of the curve needs to be optimized. The optimization method is to jump to step 1.1, describing the pile point, and adjust the control parameter ri or θi of the pile point i.

[0100] See Figure 4 As shown, in one embodiment, the radius of curvature of the original bypass path passing through the original bypass point is less than the threshold value R. min By changing r in the description of the pile points i Or θ i By changing the position of the slalom point, the trend of the slalom curve is altered, ultimately causing the radius of curvature of the slalom path to exceed the threshold value R. min ;

[0101] 1.5 Collision Optimization

[0102] According to the path description (x) k ,y k ,s k Calculate the distance D between each point on the path around the pile near pile i and pile i. k ;

[0103] D k =((x) k -x i ) 2 +(y k -y i ) 2 ) 0.5 ;

[0104] When D k Example of less than minimum safety r imin At this point, there is a risk of collision, requiring optimization to avoid it; the optimization method is to jump to step 1.1, describing the stake point, and adjust the control parameter r of stake point i. i Or θ i .

[0105] See Figure 5 As shown, in one embodiment, the distance between the original bypass path passing through the original bypass point and the original bypass point is less than a preset threshold value r. imin By changing r in the description of the pile points i Or θ iBy changing the position of the bollard point, the trend of the bollard curve is altered, ultimately ensuring that the distance between the bollard path and the bollard point exceeds the preset threshold value r. imin ;

[0106] 2. Speed ​​Planning

[0107] 2.1 Solving for the aiming point

[0108] Based on the vehicle's current position, motion state, and the path planned in step 1, the current aiming point is determined. The specific solution involves the following steps:

[0109] Step 1: Solving the pre-aiming distance:

[0110] S pre =S j +v j *T basic +0.5*a j *T basic 2 ;

[0111] in,

[0112] j = 1, 2, ... W, where j is the index of the speed planning point and W is the number of speed planning points;

[0113] S pre This represents the path length corresponding to the current target time point.

[0114] S j This represents the path length corresponding to the current vehicle's location at the waypoint.

[0115] v j The current speed of the vehicle;

[0116] a j This represents the current acceleration of the vehicle.

[0117] T basic The preset reference aiming time is a fixed value, which is 0.8s in one specific embodiment;

[0118] Step 2: Solving for the aiming point

[0119] Based on the pre-aiming distance S obtained in step one pre In path planning, based on the curve descriptions sx and sy, find S respectively. pre The corresponding x and y coordinates pre and y pre The coordinates of the vehicle's current aiming point in the geodetic coordinate system are (x... pre y pre )

[0120] Step 3: Coordinate Transformation

[0121] Establish a vehicle coordinate system with the current center of gravity of the vehicle as the origin; the longitudinal direction of the vehicle body as the x-axis, and the vehicle's direction of travel as the positive x-axis; the lateral direction of the vehicle body as the y-axis, and the left side of the vehicle's forward direction as the positive y-axis; transform the coordinates of the pre-aiming point to the vehicle coordinate system. The specific calculation method is as follows:

[0122] ΔLon=(x j -x pre )*cosΨ j +(y j -y pre )*sinΨ j ;

[0123] ΔLat=-(x j -x pre )*sinΨ j +(y j -y pre )*cosΨ j ;

[0124] in,

[0125] ΔLon is the x-coordinate of the preview point in the vehicle coordinate system;

[0126] ΔLat is the y-coordinate of the preview point in the vehicle coordinate system;

[0127] x j ,y j The coordinates of the vehicle's current position in the geodetic coordinate system;

[0128] Ψ j This represents the vehicle's current heading angle in geodetic coordinates.

[0129] The coordinates of the aiming point in the vehicle coordinate system are (ΔLon, ΔLat);

[0130] 2.2 Solving for Alternative Accelerations

[0131] Using the vehicle's current motion state and the coordinates (ΔLon, ΔLat) of the preview point obtained in section 2.1 in the vehicle coordinate system, we can calculate the vehicle's alternative lateral and longitudinal accelerations at the next moment. The specific method is as follows:

[0132] T = T basic +m*deltaT;

[0133] ax Reserve =2*(ΔLon-vx) j *T) / T 2

[0134] ay Reserve=2*(ΔLat-vy) j *T) / T 2 ;

[0135] in,

[0136] deltaT is the relaxation time, which is 0.1s in one embodiment;

[0137] m is the relaxation coefficient, which in one embodiment takes values ​​of -3, -2, -1, 0, 1, 2, 3;

[0138] T is the pre-aiming time; by using the relaxation time deltaT and the relaxation coefficient m, the pre-aiming time can be adjusted to vary around the baseline pre-aiming time Tbasic, thus obtaining multiple different pre-aiming times;

[0139] vx j This represents the x-axis component of the vehicle's current velocity v in the vehicle coordinate system.

[0140] vy j This represents the y-axis component of the vehicle's current velocity v in the vehicle coordinate system.

[0141] ax Reserve As an alternative longitudinal acceleration; ax Reserve The calculated values ​​differ depending on the aiming time.

[0142] ay Reserve As an alternative lateral acceleration; ay Reserve The calculated values ​​differ depending on the aiming time.

[0143] 2.3 Dynamic Verification

[0144] Verify the candidate longitudinal acceleration ax in section 2.2 based on the GG diagram. Reserve and alternative lateral acceleration ay Reserve Does it meet the vehicle dynamics requirements?

[0145] See Figure 7 As shown, the GG diagram is a commonly used diagram in the field of vehicles to describe the safe maneuver range of a vehicle; its x-axis represents lateral acceleration and its y-axis represents longitudinal acceleration; the boundaries of the GG diagram describe the limits of the vehicle's performance.

[0146] See Figure 7 As shown, on the GG graph, the x-coordinate is the maximum lateral acceleration ay. Max Draw a line L1 parallel to the y-axis and passing through point P1; the closed region enclosed by line L1, the boundary of the gg graph, and the ax coordinate axis is the dynamic selectable region.

[0147] The longitudinal acceleration ax in section 2.2 Reserveand alternative lateral acceleration ay Reserve In the GG graph, points falling within the dynamics-selectable region satisfy vehicle dynamics; points not falling within the dynamics-selectable region do not satisfy vehicle dynamics.

[0148] 2.4 Speed ​​Evaluation

[0149] The points falling within the dynamic selectable region in section 2.3 are evaluated according to the preset indicators; the speed evaluation can be based on passenger user experience, specific test requirements, or vehicle driving rules.

[0150] In one specific embodiment, the evaluation index is: comfort index P. comfortable Safety indicator P safe and efficiency index P efficient ;

[0151] a) Comfort indicators;

[0152] Comfort indices are mainly used to quantitatively evaluate the impact of planned speed on comfort; one example is the jerk value in the lateral and longitudinal directions.

[0153] X jerk =(ax Reserve -ax j ) / dt;

[0154] Y jerk =(ay Reserve -ay j ) / dt;

[0155] P comfortable =P cx *x jerk +P cy *y jerk ;

[0156] in,

[0157] X jerk The jerk value is for the vertical axis.

[0158] Y jerk The jerk value is for the horizontal direction;

[0159] ax j This represents the x-axis component of the vehicle's current acceleration in the vehicle coordinate system.

[0160] ay j This represents the y-axis component of the vehicle's current acceleration in the vehicle coordinate system.

[0161] dt is the time step for velocity planning, which is 0.1s in one embodiment;

[0162] P cxThe weights are preset for the vertical Jerk values;

[0163] P cy The weights are preset for the horizontal Jerk values;

[0164] b) Safety indicators;

[0165] Safety indicators are mainly used to quantitatively evaluate the danger or perceived danger of the planned speed; one example is the lateral and longitudinal acceleration values.

[0166] a Reserve＝ (ax Reserve 2 +ay Reserve 2 ) 0.5

[0167] a μ =μ*g;

[0168] P safe =a μ -a Reserve ;

[0169] in,

[0170] a Reserve As an alternative acceleration;

[0171] μ is the road surface adhesion coefficient;

[0172] g is the acceleration due to gravity;

[0173] a μ This is the maximum driving force that can be provided under the current ground adhesion conditions;

[0174] c) Efficiency indicators;

[0175] Efficiency indicators are mainly used to quantitatively evaluate the traffic efficiency of the planned speed. One example is the difference between the planned speed and the maximum speed allowed for vehicles.

[0176] Calculate the velocity v at the next moment based on the candidate longitudinal acceleration and candidate lateral acceleration. Reserve for:

[0177] v Reserve =v j +(ax Reserve 2 +ay Reserve 2 ) 0.5 *dt;

[0178] P efficient =|v maxLimit -v Reserve |;

[0179] in,

[0180] v maxLimit The maximum speed at which a vehicle is permitted to travel, such as road / environmental speed limits;

[0181] The final evaluation index P is calculated using a weighted summation method:

[0182] P = P c *P comfortable +P s *P safe +P e *P efficient ;

[0183] in,

[0184] P c The weighting of comfort indicators;

[0185] P s Weights for safety indicators;

[0186] P e The weights for efficiency indicators;

[0187] 2.5 Speed ​​Selection

[0188] The candidate longitudinal acceleration and candidate lateral acceleration with the smallest P value in selection area 2.4 are used as the final longitudinal acceleration value ax. Final and lateral acceleration value ay FInal ;

[0189] The acceleration at that point is:

[0190] ax j+1 =(ax Final 2 +ay Final 2 ) 0.5 ;

[0191] The velocity at that point is

[0192] v j+1 =v j +ax j+1 *dt;

[0193] The path length to this point is:

[0194] S j+1 =s j +v j+1 *dt;

[0195] Finally, it should be noted that the above preferred embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail through the above preferred embodiments, those skilled in the art should understand that various changes can be made to it in form and detail without departing from the scope defined by the claims of the present invention; the dimensions described in the drawings and embodiments are not related to the specific physical object and are not used to limit the protection scope of the present invention. The physical dimensions can be selected and changed according to actual needs.

Claims

1. A method for intelligent vehicle trajectory planning based on indicator tags, characterized in that, include: Path planning includes the following steps: establishing assumed around-the-piles coordinates based on pile foundation data containing pile location information and pile safety distance information; forming an assumed path based on the assumed around-the-piles coordinates; performing collision optimization on the assumed path to obtain the preferred around-the-piles point; and planning the target path passing through the preferred around-the-piles point. The path planning also includes a curvature optimization step, which is set between the path description step and the collision optimization step. Specifically: Curvature optimization - On the planned path formed in the path description step, multiple curvature feature points are taken and their curvature radii are calculated respectively. When the curvature radius of any curvature feature point is less than the threshold value of the preset driving radius when the vehicle performs a turning operation, the process jumps to the cone deflection point description step. The assumed cone deflection point is re-established with the corrected cone deflection distance and / or cone deflection angle, and path description and curvature optimization are performed until the curvature radius of each curvature feature point is greater than the threshold value. The initial planned path that meets the requirements is set as the preferred planned path. Collision optimization - On the preferred planning path, take collision feature points, calculate the expected distance between each collision feature point and the pile point. When the expected distance of any collision feature point is less than the minimum safe distance for the pile point to bypass, return to the pile point description step, reset the assumed pile point with the corrected pile distance and / or pile angle, and then perform curvature optimization and collision optimization steps until all curvature feature points and collision feature points on the preferred planning path meet the requirements, and generate the target path; Speed ​​planning includes the steps of calculating the coordinates of the aiming point (measured in reference aiming time) along the target path based on the vehicle's current operating information and correcting the vehicle's speed based on the aiming point coordinates. The velocity planning process includes the following steps when calculating acceleration: pre-aiming point solution, alternative acceleration solution, and dynamic verification. Preview point solution - Set the coordinates of the next preview point along the target path based on the preset baseline preview time and the vehicle's current operating information; Alternative acceleration solution - Based on multiple alternative aiming times with reference to the baseline aiming time, calculate multiple alternative accelerations along the target path to the next aiming point. The alternative accelerations include lateral acceleration values ​​and longitudinal acceleration values. Dynamics verification - Fill multiple candidate accelerations into the GG graph and filter the preferred acceleration points that fall into the preset dynamics region of the GG graph; If multiple preferred acceleration points are obtained from the dynamics verification, the target acceleration is obtained by selecting one according to the preset point selection rules.

2. The intelligent vehicle trajectory planning method based on indicator tags according to claim 1, characterized in that, The path planning includes the following steps: Pile point description - Identify pile points and obtain pile point location information, pile point size, and minimum safe distance for pile point detour; Description of the bypass point - The bypass direction is determined based on the location information of the bypass point, the bypass distance is set according to the minimum safe distance that needs to be maintained when bypassing the bypass point, and the assumed bypass point is established based on the bypass point data information including the bypass direction and the bypass distance. Path Description - The path between adjacent assumed stake points is set as the assumed path. The assumed path is decomposed into multiple basic paths. A horizontal line graph is plotted using the path length and x-coordinate of each basic path segment, and a vertical line graph is plotted using the path length and y-coordinate of each basic path segment. Spline interpolation is performed on the horizontal and vertical line graphs to obtain the horizontal and vertical curves, respectively. The horizontal and vertical curves are then merged to form the planned path. Collision optimization - Take multiple collision feature points on the planned path, calculate the expected distance between each collision feature point and the pile point. When the expected distance of any collision point is less than the minimum safe distance for the pile point to bypass, return to the pile point description step, reset the assumed pile point with the corrected pile distance and / or pile angle, until all collision feature points on the planned path meet the requirements, and generate the target path.

3. The intelligent vehicle trajectory planning method based on indicator tags according to claim 1, characterized in that, The preset point selection rules include the steps of speed evaluation and speed selection, specifically: Speed ​​evaluation - The assignment of values ​​to the candidate acceleration corresponding to the preferred acceleration point falling into the preset dynamic region in terms of vehicle driving comfort, safety and efficiency indicators; Speed ​​selection - The target acceleration with the best balance performance is measured by weighted average method based on the preset weights of comfort, safety and efficiency indicators, and the expected operating speed of the vehicle is measured based on the target acceleration.

4. A method for intelligent vehicle trajectory planning based on indicator tags according to claim 1 or 2, characterized in that, In the pile point description step, the reference line for the pile angle extending radially through the pile point is set with the included angle between the pile points as a parameter. The reference line for the pile angle is deflected by a preset pile angle and intersected with a circle with the pile distance as the radius and the pile point as the center to generate a hypothetical pile point.

5. A method for intelligent vehicle trajectory planning based on indicator tags according to claim 1 or 3, characterized in that, The pre-aiming point solution includes the steps of calculating the pre-aiming distance, setting the pre-aiming point coordinates based on the pre-aiming distance, and converting the pre-aiming point coordinates into the horizontal and vertical coordinates of the pre-aiming point with the vehicle as the origin.

Images