CN109540159B - Rapid and complete automatic driving track planning method - Google Patents

Rapid and complete automatic driving track planning method Download PDF

Info

Publication number
CN109540159B
CN109540159B CN201811183196.7A CN201811183196A CN109540159B CN 109540159 B CN109540159 B CN 109540159B CN 201811183196 A CN201811183196 A CN 201811183196A CN 109540159 B CN109540159 B CN 109540159B
Authority
CN
China
Prior art keywords
path
planning
expansion
curvature
turning
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN201811183196.7A
Other languages
Chinese (zh)
Other versions
CN109540159A (en
Inventor
余卓平
曾德全
熊璐
李奕姗
张培志
夏浪
卫烨
严森炜
李志强
付志强
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Tongji University
Original Assignee
Tongji University
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Tongji University filed Critical Tongji University
Priority to CN201811183196.7A priority Critical patent/CN109540159B/en
Publication of CN109540159A publication Critical patent/CN109540159A/en
Application granted granted Critical
Publication of CN109540159B publication Critical patent/CN109540159B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
    • G01C21/00—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00
    • G01C21/26—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00 specially adapted for navigation in a road network
    • G01C21/34—Route searching; Route guidance
    • G01C21/3446—Details of route searching algorithms, e.g. Dijkstra, A*, arc-flags, using precalculated routes

Landscapes

  • Engineering & Computer Science (AREA)
  • Radar, Positioning & Navigation (AREA)
  • Remote Sensing (AREA)
  • Automation & Control Theory (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Traffic Control Systems (AREA)

Abstract

The invention relates to a fast and complete automatic driving track planning method, which comprises the following steps: 1) planning a basic path; 2) performing collision detection on the basic driving path, and performing step 3), if no collision exists, judging whether the basic driving path reaches a target point, if so, performing step 4), and if not, performing step 3); 3) step 4) after obtaining the smooth effective path by adopting a sampling/searching path planning method; 4) after curvature extreme values corresponding to the paths are obtained, carrying out sectional speed planning to obtain tracks; 5) and (2) performing collision detection on the track in a time domain and a space domain, if the collision exists, judging whether the distance between the vehicle and the obstacle is more than 2 times of the minimum braking distance, if so, performing speed re-planning and returning to the step 4), and if not, returning to the step 1), and finally combining the path and the speed to generate the track and output the track.

Description

Rapid and complete automatic driving track planning method
Technical Field
The invention relates to the field of track planning of unmanned vehicles under urban structured roads, in particular to a rapid and complete automatic driving track planning method.
Background
With the development of society and the continuous improvement of the living standard of people, the number of the passenger cars in China is increased by 13.5 times in 17 years from 2000 to 2017, and the number of the passenger cars in the future is estimated to be 4 hundred million. However, the increasing energy crisis, high-load traffic pressure and social requirements for driving safety have accelerated the pace of landing unmanned technologies. As one of the cores of the unmanned technology, a trajectory planning strategy must improve the real-time performance of an algorithm to respond to the change of the environment and avoid unnecessary personal and property losses caused by collision, and meanwhile, the strategy must also improve the completeness of the algorithm to adapt to the change of traffic, improve the comfort and reduce traffic congestion.
In a traditional automatic driving track planning strategy, an algorithm with analytic completeness or probability completeness, such as a sampling method (such as a), a searching method (such as RRT) and the like, is generally considered, but the algorithm has a blind searching process, the solving process is extremely time-consuming, a generated track is a broken line segment of a segment, the curvature is not continuous, and a vehicle is difficult to execute; the fast planning methods such as a Dubins curve, a Reeds-Shepp curve and a spline curve are adopted, but the algorithms are not complete, the capability of solving to obtain a feasible solution is sharply reduced when the environment is relatively complex, and the time consumption is continuously increased. In addition, under urban structured road conditions, the result of trajectory planning must comply with traffic regulations, complying with the behavioral characteristics of the driver.
Therefore, how to provide a parking strategy and system that solve the above problems is a problem that needs to be solved by those skilled in the art.
Disclosure of Invention
The invention aims to overcome the defects of the prior art and provide a fast and complete automatic driving track planning method.
The purpose of the invention can be realized by the following technical scheme:
a fast and complete automatic driving track planning method comprises the following steps:
1) planning a basic path, and acquiring a basic driving path imitating the behavior of a driver;
2) performing collision detection on the basic driving path, if the collision exists, recording the position of the collision on the basic driving path, and performing step 3), if the collision does not exist, judging whether the basic driving path reaches a target point, if so, performing step 4), and if not, performing step 3);
3) step 4) after obtaining the smooth effective path by adopting a sampling/searching path planning method;
4) acquiring a curvature curve of the path by adopting a speed planning method, and performing sectional speed planning after acquiring a curvature extreme value corresponding to the path to obtain a track;
5) and (3) carrying out collision detection on the track in a time domain and a space domain, if no collision exists, judging whether the distance between the vehicle and the obstacle is more than 2 times of the minimum braking distance, if so, carrying out speed re-planning and returning to the step 4), otherwise, returning to the step 1), and finally combining the path and the speed to generate the track and outputting the track.
The step 1) specifically comprises the following steps:
11) obtaining vehicle key parameters including vehicle structure parameters, vehicle actuator performance parameters and road characteristic parameters, wherein the vehicle structure parameters include wheelbase, wheel base, vehicle length, vehicle width, vehicle weight and centroid position, the vehicle actuator performance parameters include maximum vehicle speed, minimum vehicle speed, maximum acceleration and minimum acceleration, and the road characteristic parameters include road adhesion coefficient and road roughness;
12) calculating a curvature limit value according to the vehicle key parameters, wherein the curvature limit value is the minimum value of the curvature of the minimum turning radius limit value, the maximum curvature of the lateral maximum lateral acceleration limit value and the maximum curvature of the road adhesion coefficient limit value;
13) judging whether to carry out high-comfort track planning according to user requirements or self-adaptive decisions:
131) when the comfort requirement is not high, namely the road speed limit is not more than 20km/h, the control points of the simulated driver behavior path guided by four driving behaviors of straight running, lane changing/merging, turning and turning are respectively obtained, and then a B spline curve or a Bessel curve is used for fitting the control points to finally obtain a smooth basic driving path;
132) when the requirement on comfort is high, namely the road speed limit is more than 20km/h, two driving behavior guiding path control points of turning and turning are respectively obtained, then the control points are fitted by using the spiral line segments and the circular arc path segments, and finally a smooth basic driving path is obtained.
The step 3) specifically comprises the following steps:
31) cutting the basic driving path to obtain an effective path;
32) updating the effective path or the effective node into a path tree, wherein the path tree is a dynamic kd tree, and the position adjustment of the nodes in the tree realizes the comprehensive normalized Euclidean distance and the accumulated course angle deviation;
33) judging whether the number of times of node expansion of sampling/searching reaches a set threshold value, if so, performing step 37), and if not, performing step 34);
34) selecting a sampling method or a searching method expansion node;
35) collision detection is carried out on the expansion node and the connecting line between the expansion node and the father node, if collision exists, the node is not stored in the path tree, and the step 33 is returned;
36) judging whether the node is expanded to the target point, if not, returning to the step 32), and if so, carrying out 37);
37) and (4) taking out effective nodes in the path tree, and fitting the control points by using a B-spline curve or a Bessel curve to obtain a smooth effective path.
In the step 131), a control point imitating the driver behavior path is obtained according to the curvature limit value, and the control point meets the following constraints:
Figure GDA0002619243030000031
wherein L is the Euclidean distance between two control points, alpha is the included angle between the control points, and klimitIs the curvature limit.
In the step 131) described above, the step,
the straight-going path guided by the driving behavior is composed of n1 control straight-line segments, and the straight-line segments meet the following constraints:
Figure GDA0002619243030000032
Figure GDA0002619243030000033
wherein lend-start1For Euclidean distance from the planning end point to the planning start point, li1Is the length of the i1 straight line segment, alphai1Is the included angle between the control points of the straight line segment of the i1 th segment;
the lane changing/merging road path determined according to the driving behavior is composed of n2 control straight line segments, and the straight line segments meet the following constraints:
Figure GDA0002619243030000034
Figure GDA0002619243030000041
wherein lend-start2For Euclidean distance from the planning end point to the planning start point, li2Is the length of the i2 straight line segment, alphai2Is the included angle between the control points of the straight line segment of the i2 th segment;
the turning path determined according to the driving behavior is composed of n3 control straight line segments, and the straight line segments meet the following constraints:
turning restraint without lane borrowing:
Figure GDA0002619243030000042
Figure GDA0002619243030000043
Figure GDA0002619243030000044
Figure GDA0002619243030000045
turning restraint of the left lane borrowing:
Figure GDA0002619243030000046
Figure GDA0002619243030000047
Figure GDA0002619243030000048
Figure GDA0002619243030000049
turning restraint of upward lane borrowing:
Figure GDA0002619243030000051
Figure GDA0002619243030000052
Figure GDA0002619243030000053
Figure GDA0002619243030000054
wherein, thetaj3Is the starting course angle, θg3Is the end point course angle, xg3、yg3As end point coordinate, xj3、yj3As coordinates of the starting point,/i3Is the length of the i3 straight line segment, alphai3Is the included angle between the control points of the straight line segment of the i3 th segment;
the turn-around path determined according to the driving behavior is composed of n4 control straight line segments, and the straight line segments meet the following constraints:
turning restraint without lane borrowing:
Figure GDA0002619243030000055
Figure GDA0002619243030000056
Figure GDA0002619243030000057
Figure GDA0002619243030000058
turning restraint of the left lane borrowing:
Figure GDA0002619243030000061
Figure GDA0002619243030000062
Figure GDA0002619243030000063
Figure GDA0002619243030000064
turning constraint of right lane borrowing:
Figure GDA0002619243030000065
Figure GDA0002619243030000066
Figure GDA0002619243030000067
Figure GDA0002619243030000068
turning restriction of lane borrowing on the left and the right:
Figure GDA0002619243030000069
Figure GDA00026192430300000610
Figure GDA00026192430300000611
Figure GDA00026192430300000612
wherein, thetaj4Is the starting course angle, θg4Is the end point course angle, xg4、yg4As end point coordinate, xj4、yj4As coordinates of the starting point,/i4Is the length of the i4 straight line segment, alphai4Is the included angle between the control points of the straight line segment of the i4 th segment.
In the step 132), a section of arc and two sections of spiral lines are adopted according to the improved turning and turning path of the driving behavior, so that the following constraints are met:
the first helix:
a1s2+b1s+c1＝k1
c1＝kstart
Figure GDA0002619243030000071
Figure GDA0002619243030000072
Figure GDA0002619243030000073
the second helix:
a2s2+b2s+c2＝k2
c2＝kend
Figure GDA0002619243030000074
Figure GDA0002619243030000075
Figure GDA0002619243030000076
circular arc curve:
x1＝Rx-Rsinθ1
y＝Ry+Rcosθ1
x2＝Rx-Rsinθ2
y2＝Ry+Rsinθ2
where s is the path length, s1Is the length of the first helical line, s2Is the length of the second helical line, a1、b1、c1For the first spiral segment, a2、b2、c2For the second spiral, a characteristic parameter, k, is to be determined1Is the first section of the helix curvature, k2Is the curvature of the second spiral, kstartAs the curvature of origin, kendTo end point curvature, θiAs starting point heading, θendFor end course, θ1Is the end course of the first spiral line, theta2The starting course of the second spiral line, R is the radius of the circular arc, Rx，RyIs the center coordinate of the arc, x1，y1Coordinates of the start of the arc, x2，y2And the arc terminal point coordinates and the integral of the spiral line are solved by adopting a Simpson formula, so that the calculation complexity is reduced, and the real-time performance of the algorithm is improved.
In the step 31), the path cutting specifically comprises:
and backtracking to a nearest non-collision control point along the basic path through the collision point detected by collision, wherein the non-collision control point is used as a starting point of the sampling/searching path plan, and a smooth path from the non-collision control point to the starting point of the basic path plan is used as an effective path obtained by cutting.
In the step 3), the switching mode of the sampling/searching path planning comprises user setting and self-adaptive selection according to the complexity of the environment by a track planning strategy, when more obstacles exist in the environment or the environment map is larger, the RRT is adopted for rapid expansion, and when the environment map is simpler or smaller, the A is adopted for heuristic searching.
In the step 34) described above, the step,
the sampling method is an improved RRT method, the node expansion modes of the method comprise three modes of forced expansion, driving behavior bias expansion and random expansion, and the specific self-adaptive selection methods of the three expansion modes are as follows:
when the environment is single and the success rate is higher than a set threshold value, adopting forced expansion; when the success rate of the multiple forced expansion is smaller than a set threshold value, adopting driving behavior bias expansion; when the success rate is low and the environment complexity is high, a standard expansion mode is adopted;
the forced expansion is to directly store a target point as an expanded node into a path tree, to perform collision detection to verify the validity of the expanded node, to keep the expanded node if the expanded node is valid, and to delete the expanded node if the expanded node is invalid, the driving behavior bias expansion firstly labels an expanded domain with a sector according to the driving behavior and generates a relay node set, an intersection node set and a target node set at appropriate positions, a seed point is selected from the sector labeled domain with a Gaussian random form, the node is stored into the path tree, to perform collision detection to verify the validity of the expanded node, to keep the expanded node if the expanded node is valid, to delete the expanded node set if the expanded node is invalid, the standard expansion firstly labels the expanded domain with a regular polygon spliced by rectangles or rectangles according to the driving behavior and generates the relay node set, the intersection node set and the target node set at appropriate positions, and the seed point is selected from the, and storing the nodes into the path tree, performing collision detection, verifying the validity of the expansion nodes, and if the validity is valid, keeping the validity, and if the invalidity is invalid, deleting the validity.
The searching method adopts an improved A method, the node expansion mode comprises three modes of driving behavior bias searching, steering angle constraint searching and standard searching, and the specific self-adaptive selection methods of the three expansion modes are as follows:
when the environment is single and the success rate is higher than a set threshold value, expanding the driving behavior tendency search; when the success rate of the driving behavior tendency search for many times is smaller than a set threshold value, adopting the expansion of steering angle constraint search; and when the success rate is low and the environment complexity is high, adopting an expansion mode of standard search.
The expansion of the driving behavior bias search is to label an expanded domain by a sector, generate a relay node set, an intersection node set and a target node set at proper positions, carry out node expansion in the directions of 220 path bases which meet the minimum turning radius constraint of a vehicle, store the expanded nodes into a path tree, carry out collision detection, verify the validity of the expanded nodes, retain the expanded nodes if the expanded nodes are valid, delete the expanded nodes if the expanded nodes are invalid, constrain the expansion of the search by the steering angle, label the expanded domain by a regular polygon spliced by rectangles or rectangles, generate the relay node set, the intersection node set and the target node set at proper positions, carry out node expansion in the directions of 220 path bases which meet the minimum turning radius constraint of the vehicle, store the expanded nodes into the path tree, carry out collision detection, verify the validity of the expanded nodes, and retain the expanded nodes if the expanded nodes are valid, and deleting the expanded nodes if the expanded nodes are invalid, expanding the standard search, marking the expanded domain by using regular polygons spliced by rectangles or rectangles, generating a relay node set, an intersection node set and a target node set at proper positions, expanding the nodes along the directions of 20 path bases in the vertical direction and the longitudinal direction, storing the expanded nodes into a path tree, performing collision detection, verifying the validity of the expanded nodes, and if the expanded nodes are valid, keeping the expanded nodes, and if the expanded nodes are invalid, deleting the expanded nodes.
In the step 4), the curvature extreme value subsection speed planning carries out the uniform speed planning between two curvature extreme points of the basic path, the acceleration and deceleration speed planning which is restricted by the road surface attachment and the actuator performance is adopted outside the extreme points,
or constant speed planning is carried out between the front point and the rear point of the four curvature extreme points of the basic path and the sampling/searching path, and acceleration and deceleration speed planning which is restricted by road surface adhesion and actuator performance is adopted outside the extreme points.
The switching mode of the sampling/searching path planning can be set by a user, and can also be self-adaptively selected by a track planning strategy according to the complexity of the environment, and when a plurality of obstacles exist in the environment or the environment map is large, the RRT is adopted for rapid expansion; and when the environment is simpler or the environment map is smaller, performing heuristic search by adopting A.
In the step 5), the time domain and space domain collision detection is to represent the time domain by speed, construct a three-dimensional coordinate system by representing the space domain by a path, and respectively use the detection range and the road boundary of a sensor as the upper limit and the lower limit of the space domain, and use the road speed limit, the road surface condition speed limit and the vehicle actuator performance speed limit as the upper limit and the lower limit of the time domain;
in the step 5), the re-planning decision is specifically to decide whether to give priority to speed planning or not by combining the environmental complexity and the traffic rules on the basis of whether the collision point and the collision time are in the safe planned domain or not.
Compared with the prior art, the invention has the following advantages:
the invention provides a driving behavior guiding track planning method, which is suitable for urban structured roads, reduces blindness of a planning process, reduces time consumption of track planning, generates a track according with behavior characteristics of drivers, and makes an automatic driving process more stable and comfortable.
Firstly, the planned track comprehensively considers vehicle dynamics constraint, vehicle actuator performance constraint and road surface conditions, a B spline or a Bessel curve is adopted to smooth the track, and the generated track has continuous curvature and strong performability.
The invention designs a basic path planning method imitating the behavior of a driver, which divides the driving behavior into straight movement, lane changing/merging, turning and turning, so that the planning process is simpler, the algorithm complexity is reduced, and the real-time performance is high.
The invention designs an improved sampling planning and improved searching method, and the improved sampling planning and improved searching method is complementary with the basic path method, so that the track planning method has completeness, and the track planning result is reliable and stable.
The invention designs a curvature extreme value segmented speed planning method which is decoupled with path planning to realize modularization of the path planning and the speed planning, and meanwhile, the speed planning of the front path and the rear path has portability, the complexity of a speed planning algorithm is low, and the real-time performance is strong.
In the specific implementation process of the trajectory planning, the complexity of the environment, the vehicle state, the driver requirement and the algorithm success rate are considered, a plurality of self-adaptive links are set for the algorithm, and the strategy has strong adaptability to the environment and the driver requirement.
Drawings
FIG. 1 is a flow chart of an automatic driving trajectory planning strategy with rapidity and completeness according to the present invention;
FIG. 2 is a flowchart illustrating a specific implementation of the trajectory planning strategy of the present invention in which a driving behavior-oriented path is merged with a change RRT;
FIG. 3 is a flowchart illustrating an embodiment of a trajectory planning strategy of the present invention incorporating a driving behavior-oriented path and an improvement A;
fig. 4 is a schematic diagram of the lane change/merge driving behavior oriented path control points and the smoothed path of the present invention. Wherein, fig. 4a is a schematic diagram of a control straight-line segment of a path control point structure guided by lane changing/merging driving behaviors, and fig. 4B is a schematic diagram of the control straight-line segment of the path control point structure guided by lane changing/merging driving behaviors after being smoothed by a B spline curve;
FIG. 5 is a schematic diagram of turning driving behavior oriented path control points and a smoothed path in accordance with the present invention. Wherein, fig. 5a is a schematic diagram of a control straight-line segment of a path control point structure guided by no-way turning driving behavior, and fig. 5B is a schematic diagram of a control straight-line segment of a path control point structure guided by no-way turning driving behavior after being smoothed by a B-spline curve;
FIG. 6 is a schematic diagram of a turning behavior oriented path control point and a smoothed path of the present invention. Wherein, fig. 6a is a schematic diagram of a control straight-line segment of the path control point structure guided by the left lane-borrowing turning driving behavior, and fig. 6B is a schematic diagram of the control straight-line segment of the path control point structure guided by the left lane-borrowing turning driving behavior after being smoothed by a B-spline curve;
FIG. 7 is a schematic diagram of turning driving behavior oriented path control points and a smoothed path in accordance with the present invention. Wherein, fig. 7a is a schematic diagram of a control straight-line segment of the path control point structure guided by the upper lane-borrowing turning driving behavior, and fig. 7B is a schematic diagram of the control straight-line segment of the path control point structure guided by the upper lane-borrowing turning driving behavior after being smoothed by a B-spline curve;
fig. 8 is a schematic diagram of a turn-around driving behavior oriented path control point and a smoothed path according to the present invention. Wherein, fig. 8a is a schematic diagram of a control straight-line segment of a path control point structure guided by the non-lane-turning driving behavior, and fig. 8B is a schematic diagram of a control straight-line segment of a path control point structure guided by the non-lane-turning driving behavior after being smoothed by a B-spline curve;
fig. 9 is a schematic diagram of a turn-around driving behavior oriented path control point and a smoothed path according to the present invention. Wherein, fig. 9a is a schematic diagram of a control straight-line segment of the structure of the path control point guided by the left lane change driving behavior, and fig. 9B is a schematic diagram of the control straight-line segment of the structure of the path control point guided by the left lane change driving behavior after being smoothed by a B-spline curve;
fig. 10 is a schematic diagram of the turn-around driving behavior oriented path control points and the smoothed path according to the present invention. Wherein, fig. 10a is a schematic diagram of a control straight-line segment of a path control point structure guided by right lane change driving behavior, and fig. 10B is a schematic diagram of a control straight-line segment of the path control point structure guided by right lane change driving behavior after being smoothed by a B spline curve;
FIG. 11 is a schematic view of a turn-around driving behavior oriented path control point and a smoothed path in accordance with the present invention; wherein, fig. 11a is a schematic diagram of a control straight-line segment of a path control point structure guided by the lane-changing driving behavior on the left and right, and fig. 11B is a schematic diagram of a control straight-line segment of a path control point structure guided by the lane-changing driving behavior on the left and right after being smoothed by a B spline curve;
FIG. 12 is a schematic of the smooth path of an improved turn and turn driving behavior guide of the present invention. Wherein FIG. 12a is a smooth path diagram of an improved cornering driving behaviour guide and FIG. 12b is a smooth path diagram of an improved cornering driving behaviour guide;
FIG. 13 is a schematic diagram of a lane change/merge driving behavior-oriented path cutting to obtain an effective path according to the present invention;
fig. 14 is a schematic diagram of three sampling/search spaces oriented by the straight-line and lane-change/lane-merging driving behaviors of the present invention. Wherein, fig. 14a is a schematic view of a scene of driving behaviors of straight-going and lane-changing/lane-merging, fig. 14b is a schematic view of random expansion/standard search of the driving behaviors of straight-going and lane-changing/lane-merging, fig. 14c is a schematic view of bias expansion/corner constraint search/bias search of the driving behaviors of straight-going and lane-changing/lane-merging, and fig. 14d is a schematic view of forced expansion of the driving behaviors of straight-going and lane-changing/lane-merging;
FIG. 15 is a schematic view of the present invention in a cornering situation.
FIG. 16 is a schematic representation of a sampling/search space for steering cornering behaviour according to the invention;
FIG. 17 is a schematic view of a sampling/search space directed to cornering behaviour according to the invention;
FIG. 18 is a schematic illustration of a turn-around driving behavior-oriented environment of the present invention;
FIG. 19 is a schematic view of a sample/search space directed toward turn-around driving behavior in accordance with the present invention;
FIG. 20 is a schematic view of a sample/search space directed toward turn-around driving behavior in accordance with the present invention;
FIG. 21 is a schematic diagram of a node expansion according to the present invention;
FIG. 22 is a schematic diagram of an expanded node according to the present invention;
FIG. 23 is a schematic diagram of a node expansion according to the present invention;
FIG. 24 is a schematic view of a path curvature of the present invention;
FIG. 25 is a schematic view of a velocity profile of the present invention;
FIG. 26 is a schematic view of a path curvature of the present invention;
FIG. 27 is a schematic view of a velocity profile of the present invention;
FIG. 28 is a schematic diagram of a velocity profile of the present invention.
Detailed Description
The invention is described in detail below with reference to the figures and specific embodiments.
Examples
Fig. 1 is a flowchart of an automatic driving trajectory planning strategy with rapidity and completeness according to the present invention, and fig. 2 and 3 are flowcharts of two different implementations of the automatic driving trajectory planning strategy with rapidity and completeness according to the present invention. The method comprises the following specific steps:
step 1: as shown in fig. 1, vehicle key parameters including vehicle structural parameters, vehicle actuator performance parameters, and road characteristic parameters are obtained. The vehicle structure parameters comprise wheelbase, wheel base, vehicle length, vehicle width, vehicle weight, mass center position and the like, the performance parameters of the vehicle execution actuator comprise maximum vehicle speed, minimum vehicle speed, maximum acceleration, minimum acceleration and the like, and the road characteristic parameters comprise road adhesion coefficient, road roughness and the like;
step 2: as shown in fig. 1, calculating a curvature limit from the vehicle key parameters, where the curvature limit is min { curvature of minimum turning radius limit, maximum curvature of lateral maximum lateral acceleration limit, maximum curvature of road adhesion coefficient limit };
and step 3: as shown in fig. 1, according to the requirement of the user, or with the current vehicle speed, the road curvature change, etc. as the adaptive conditions, whether high-comfort trajectory planning is needed is selected, if yes, the procedure goes to step 31, otherwise, the procedure goes to step 32:
step 31: respectively solving path control points for guiding four driving behaviors of straight driving, lane changing/merging, turning and turning;
step 311: calculating a control point according to the curvature limit value in the step 2, wherein the constraint equation which needs to be met by the control point is
Figure GDA0002619243030000131
Wherein L is the Euclidean distance between two control points, alpha is the included angle between the control points, and klimitIs the curvature limit. Respectively turning to step 312, step 313, step 314 or step 315 according to the driving behavior decision;
step 312: the straight-line path is composed of n1 segment length control straight-line segments, and the constraint equation required to be satisfied by the straight-line segments is as follows:
Figure GDA0002619243030000132
Figure GDA0002619243030000133
wherein lend-start1Is the Euclidean distance from the planning end point to the planning start point, li1Is the length of the i1 straight line segment, alphai1The included angle between the control points of the straight line segment of the i1 th segment.
Step 313: as shown in fig. 4, the lane change/merging road path is composed of n2 segment-length control straight segments, and the constraint equation that the straight segments need to satisfy is as follows:
Figure GDA0002619243030000134
Figure GDA0002619243030000135
wherein lend-start2Is the Euclidean distance from the planning end point to the planning start point, li2Is the length of the i2 straight line segment, alphai2The included angle between the control points of the straight line segment of the i2 th segment.
Step 314: as shown in fig. 5-7, the turning path is composed of n3 segment-length control straight-line segments, and the constraint equation that the straight-line segments need to satisfy is as follows:
turning restraint without lane borrowing:
Figure GDA0002619243030000141
Figure GDA0002619243030000142
Figure GDA0002619243030000143
Figure GDA0002619243030000144
turning restraint of the left lane borrowing:
Figure GDA0002619243030000145
Figure GDA0002619243030000146
Figure GDA0002619243030000147
Figure GDA0002619243030000148
turning restraint of upward lane borrowing:
Figure GDA0002619243030000149
Figure GDA00026192430300001410
Figure GDA00026192430300001411
Figure GDA00026192430300001412
wherein, thetaj3Is the starting course angle, θg3Is the end point course angle, xg3、yg3As end point coordinate, xj3、yj3As coordinates of the starting point,/i3Is the length of the i3 straight line segment, alphai3The included angle between the control points of the i3 th straight line segment;
step 315: as shown in fig. 8 to 11, the turnaround path is composed of n segment lengths of control straight-line segments, and the constraint equation that the straight-line segments need to satisfy is as follows:
turning restriction without borrowing lane:
Figure GDA0002619243030000151
Figure GDA0002619243030000152
Figure GDA0002619243030000153
Figure GDA0002619243030000154
turning restriction of left lane borrowing:
Figure GDA0002619243030000155
Figure GDA0002619243030000156
Figure GDA0002619243030000157
Figure GDA0002619243030000158
turning restriction of right lane borrowing:
Figure GDA0002619243030000161
Figure GDA0002619243030000162
Figure GDA0002619243030000163
Figure GDA0002619243030000164
turning restriction on both left and right sides by means of lane turning:
Figure GDA0002619243030000165
Figure GDA0002619243030000166
Figure GDA0002619243030000167
Figure GDA0002619243030000168
wherein, thetaj4Is the starting course angle, θg4Is the end point course angle, xg4、yg4As end point coordinate, xj4、yj4As coordinates of the starting point,/i4Is the length of the i4 straight line segment, alphai4The included angle between the control points of the straight line segment of the i4 th segment.
Step 32: when the comfort requirement is high, switching to the step, respectively solving the path control points for guiding the turning and turning behaviors according to the behavior decision result, and fitting the control points by using the spiral line segments and the circular arc path segments to finally obtain a smooth basic driving path;
step 321: as shown in fig. 12, the improved turning and turning path is composed of a circular arc and two spiral lines, and the constraint equation to be satisfied is as follows:
the first helix
a1s2+b1s+c1＝k1
c1＝kstart
Figure GDA0002619243030000169
Figure GDA0002619243030000171
Figure GDA0002619243030000172
Second helix
a2s2+b2s+c2＝k2
c2＝kend
Figure GDA0002619243030000173
Figure GDA0002619243030000174
Figure GDA0002619243030000175
Curve of circular arc
x1＝Rx-Rsinθ1
y＝Ry+Rcosθ1
x2＝Rx-Rsinθ2
y2＝Ry+Rsinθ2
Where s is the path length, s1Is the length of the first helical line, s2Is the length of the second helical line, a1，b1，c1For the first spiral segment, a2，b2，c2For the second spiral, a characteristic parameter, k, is to be determined1Is the first section of the helix curvature, k2Is the curvature of the second spiral, kstartAs a starting pointCurvature, kendIn order to be the end point curvature,ias the heading of the starting point,endin order to be the heading of the end point,1the heading of the terminal point of the first spiral line segment,2the starting course of the second spiral line, R is the radius of the circular arc, Rx，RyIs the center coordinate of the arc, x1，y1Coordinates of the start of the arc, x2，y2And (4) coordinates of the end point of the arc.
Step 322: the integral of the spiral line is solved by adopting a Simpson formula so as to reduce the calculation complexity and improve the real-time performance of the algorithm.
And 4, step 4: and (6) carrying out collision detection on the basic driving path, determining whether a collision exists, recording the position of the collision on the basic driving path if the collision exists, and then jumping to the step 6. (ii) a
And 5: judging whether the basic driving path reaches a target point, and jumping to the step 13 if the basic driving path reaches the target point;
step 6: as shown in fig. 13, the basic driving route is cut to obtain an effective route. The path cutting is to trace back to a nearest non-collision control point along the basic path through collision points detected by collision, the non-collision control point is used as a starting point of sampling/searching path planning, and a smooth path between the non-collision control point and the starting point of the basic path planning is used as an effective path obtained by cutting;
and 7: and updating the obtained effective path or the effective node into the path tree. The path tree is a dynamic kd tree, and the position adjustment of nodes in the tree realizes the comprehensive normalized Euclidean distance and the accumulated course angle deviation;
and 8: judging whether the expansion times of the sampled/searched nodes reach a set threshold value or not, if so, jumping to the step 12;
and step 9: and selecting a sampling method or a searching method expansion node. The switching mode of the sampling/searching path planning can be set by a user, or can be self-adaptively selected according to the complexity of the environment by a track planning strategy, and when a plurality of obstacles exist in the environment or the environment map is large, the RRT is adopted for rapid expansion, and the step 91 is executed; when the environment has a simpler or smaller environment map, performing heuristic search by adopting A, and jumping to the step 92;
step 91: the sampling method is an improved RRT, and the node expansion modes of the sampling method comprise three modes of forced expansion, driving behavior bias expansion and random expansion:
the improved RRT node expansion mode selection is self-adaptive, the success rate of node expansion, the environmental complexity and the random seed probability factors are integrated, and forced expansion is adopted when the environment is single and the success rate is higher than a certain threshold; when the success rate of the multiple forced expansion is smaller than a certain threshold value, adopting driving behavior bias expansion; and when the success rate is low and the environment complexity is high, a standard extension mode is adopted.
And the forced expansion is to directly store the target point as an expanded node into the path tree, perform collision detection, verify the validity of the expanded node, retain the node if the node is valid, and delete the node if the node is invalid.
The driving behavior bias expansion firstly labels an expansion domain with a sector according to the driving behavior, as shown in fig. 14, fig. 16 and fig. 19, and generates a relay node set, an intersection node set and a target node set at appropriate positions, seed points are selected from the sector labeling domain in a gaussian random form, nodes are stored in a path tree, collision detection is carried out, the validity of the expansion nodes is verified, if the expansion nodes are valid, the nodes are retained, and if the expansion nodes are invalid, the nodes are deleted.
According to the standard expansion, firstly, according to the driving behavior, an expansion domain is marked by regular polygons which are spliced by rectangles or rectangles, as shown in fig. 14, fig. 17 and fig. 20, a relay node set, an intersection node set and a target node set are generated at proper positions, seed points are selected from a marked domain in a standard random mode, nodes are stored in a path tree, collision detection is carried out, the validity of the expansion nodes is verified, if the expansion nodes are valid, the nodes are reserved, and if the expansion nodes are invalid, the nodes are deleted.
And step 92: the searching method is an improved A, and the node expansion mode comprises three modes of driving behavior bias searching, steering angle constraint searching and standard searching:
the improved A-node expansion mode selection is self-adaptive, integrates the cumulative success rate of node expansion and the environment complexity factor, and adopts the expansion of driving behavior bias search when the environment is single and the success rate is higher than a certain threshold; when the success rate of the driving behavior tendency search for many times is smaller than a certain threshold value, adopting the expansion of the steering angle constraint search; and when the success rate is low and the environment complexity is high, adopting an expansion mode of standard search.
Expanding the driving behavior bias search, marking an expanded domain by sectors as shown in fig. 14, 16 and 19, generating a relay node set, an intersection node set and a target node set at appropriate positions, wherein the node expansion is performed in directions of 220 path bases to meet the minimum turning radius constraint of a vehicle, storing expanded nodes into a path tree as shown in fig. 21-23, performing collision detection, verifying the validity of the expanded nodes, and if the expanded nodes are valid, keeping the expanded nodes, and if the expanded nodes are invalid, deleting the expanded nodes.
And (3) expanding the steering angle constraint search, namely labeling an expanded domain by using a regular polygon spliced by rectangles or rectangles, as shown in fig. 14-20, generating a relay node set, an intersection node set and a target node set at proper positions, expanding the nodes to meet the minimum turning radius constraint of the vehicle, and carrying out the node expansion in directions of 220 path bases, as shown in fig. 21-23, storing the expanded nodes into a path tree, carrying out collision detection, verifying the validity of the expanded nodes, if the expanded nodes are valid, keeping the expanded nodes, and if the expanded nodes are invalid, deleting the expanded nodes.
And (3) standard search expansion, namely labeling an expanded domain by using a regular polygon spliced by rectangles or rectangles, as shown in fig. 14-20, generating a relay node set, an intersection node set and a target node set at appropriate positions, performing node expansion along the directions of 20 path bases in the vertical direction and the longitudinal direction, as shown in fig. 21, storing expanded nodes into a path tree, performing collision detection, verifying the validity of the expanded nodes, and if the expanded nodes are valid, keeping the expanded nodes, and if the expanded nodes are invalid, deleting the expanded nodes.
Step 10: collision detection is carried out on the expansion node and the connecting line between the expansion node and the father node, when collision exists, the node is not stored in the path tree, and the step 8 is skipped;
step 11: judging whether the node is expanded to a target point or not, and if not, jumping to the step 7;
step 12: taking out effective nodes in the path tree, and fitting the control points by using a B spline curve or a Bezier curve to finally obtain a smooth effective path;
step 13: solving the curvature curve of the path, finding the curvature extreme values (maximum value and minimum value) of the path, and then carrying out sectional speed planning to obtain the track:
as shown in fig. 24-25, the curvature extremum piecewise velocity planning performs a constant velocity planning between two curvature extremum points of the basic path, and adopts an acceleration and deceleration velocity planning subject to the constraints of road surface adhesion and the constraints of actuator performance outside the extremum points;
as shown in fig. 26 to 28, the curvature extremum segmented speed planning is to perform a constant speed planning between two points in front of and behind four curvature extremum points of the basic path and the sampled/searched path, and an acceleration/deceleration speed planning subject to the road surface adhesion constraint and the actuator performance constraint is adopted outside the extremum points.
Step 14: the trajectory is subjected to collision detection in the time domain and the spatial domain, and if there is no collision, the step 16 is skipped. The time domain and space domain collision detection is to represent the time domain by speed, construct a three-dimensional coordinate system by representing the space domain by a path, and respectively use the detection range of a sensor and a road boundary as upper and lower limits of the space domain, and use the road speed limit, the road surface condition speed limit and the vehicle actuator performance speed limit as upper and lower limits of the time domain;
step 15: and (4) performing re-planning decision, and returning to the step (13) if speed planning is performed, or returning to the step (2) if speed re-planning is not performed. The re-planning decision is to decide whether to give priority to the speed planning or not by combining the environmental complexity and the traffic rules on the basis of whether the collision point and the collision time are in the safe planned domain or not on the basis of the step 14. (ii) a
Step 16: and combining the path and the speed into a track and outputting the track.
The invention provides an automatic driving track planning strategy with rapidity and completeness, which is characterized in that a basic path planning strategy, a sampling/searching path planning strategy, a collision detection strategy, a speed planning strategy and a re-planning strategy are organically combined by taking the behavior of a driver as guidance. The basic path planning strategy and the sampling/searching path planning strategy both take driving behaviors as guidance, so that the acceleration of the algorithm is realized, meanwhile, the completeness of the algorithm is guaranteed by the organic combination of the strategies, and the strategy can output an effective track under the condition that the path exists. The basic path planning strategy firstly needs to acquire key parameters (vehicle structure parameters, vehicle actuator performance parameters and road characteristic parameters); secondly, calculating a path curvature limit value; then selecting a basic path plan of driving behavior guidance or an improved basic path plan of driving behavior guidance according to the comfort requirement; and finally outputting a basic driving path. The sampling/searching path planning strategy is a supplementary strategy when the basic path planning strategy is difficult to meet the requirements (collision detection does not pass or the path does not reach a target point), firstly, the basic driving path is cut to obtain an effective path and a path tree is updated; secondly, carrying out sampling/searching node expansion; then, collision detection and stop detection (reaching a target point or planning time limit) are performed in sequence; finally, the valid partial path is smoothed. The speed planning strategy adopts curvature extremum segmentation speed planning. The re-planning decision strategy is used for deciding whether to perform path re-planning or speed re-planning. The invention uses the driving behavior to guide basic path planning and integrates a sampling/searching planning strategy, thereby greatly improving the real-time performance of track planning and the completeness of an algorithm, generating a track which accords with the characteristics of a human driver, and enabling the automatic driving process to be more stable and reliable.
The above description is only of the preferred embodiments of the present invention, and it should be noted that: it will be apparent to those skilled in the art that various modifications and adaptations can be made without departing from the principles of the invention and these are intended to be within the scope of the invention.

Claims (10)

1. A fast and complete automatic driving track planning method is characterized by comprising the following steps:
1) planning a basic path, and acquiring a basic driving path imitating the behavior of a driver;
2) performing collision detection on the basic driving path, if the collision exists, recording the position of the collision on the basic driving path, and performing step 3), if the collision does not exist, judging whether the basic driving path reaches a target point, if so, performing step 4), and if not, performing step 3);
3) step 4) after obtaining the smooth effective path by adopting a sampling/searching path planning method;
4) acquiring a curvature curve of the path by adopting a speed planning method, and performing sectional speed planning after acquiring a curvature extreme value corresponding to the path to obtain a track;
5) and (3) carrying out collision detection on the track in a time domain and a space domain, if the track is collided, judging whether the distance between the vehicle and the obstacle is more than 2 times of the minimum braking distance, if so, carrying out speed re-planning and returning to the step 4), and if not, returning to the step 1), and finally combining the path and the speed to generate the track and outputting the track.
2. The method for planning a fast and complete automatic driving trajectory according to claim 1, wherein the step 1) specifically comprises the following steps:
11) obtaining vehicle key parameters including vehicle structure parameters, vehicle actuator performance parameters and road characteristic parameters, wherein the vehicle structure parameters include wheelbase, wheel base, vehicle length, vehicle width, vehicle weight and centroid position, the vehicle actuator performance parameters include maximum vehicle speed, minimum vehicle speed, maximum acceleration and minimum acceleration, and the road characteristic parameters include road adhesion coefficient and road roughness;
12) calculating a curvature limit value according to the vehicle key parameters, wherein the curvature limit value is the minimum value of the curvature of the minimum turning radius limit value, the maximum curvature of the lateral maximum lateral acceleration limit value and the maximum curvature of the road adhesion coefficient limit value;
13) judging whether to carry out high-comfort track planning according to user requirements or self-adaptive decisions:
131) when the comfort requirement is not high, namely the road speed limit is not more than 20km/h, the control points of the simulated driver behavior path guided by four driving behaviors of straight running, lane changing/merging, turning and turning are respectively obtained, and then a B spline curve or a Bessel curve is used for fitting the control points to finally obtain a smooth basic driving path;
132) when the requirement on comfort is high, namely the road speed limit is more than 20km/h, two driving behavior guiding path control points of turning and turning are respectively obtained, then the control points are fitted by using the spiral line segments and the circular arc path segments, and finally a smooth basic driving path is obtained.
3. The method for planning a fast and complete automatic driving trajectory according to claim 1, wherein the step 3) specifically comprises the following steps:
31) cutting the basic driving path to obtain an effective path;
32) updating an effective path or an effective node into a path tree, wherein the path tree is a dynamic kd tree;
33) judging whether the number of times of node expansion of sampling/searching reaches a set threshold value, if so, performing step 37), and if not, performing step 34);
34) selecting a sampling method or a searching method expansion node;
35) collision detection is carried out on the expansion node and the connecting line between the expansion node and the father node, if collision exists, the node is not stored in the path tree, and the step 33 is returned;
36) judging whether the node is expanded to the target point, if not, returning to the step 32), and if so, carrying out 37);
37) and (4) taking out effective nodes in the path tree, and fitting the control points by using a B-spline curve or a Bessel curve to obtain a smooth effective path.
4. The method as claimed in claim 2, wherein in the step 131), the control points simulating the driver behavior path are obtained according to the curvature limit, and the control points satisfy the following constraints:
Figure FDA0002619243020000021
wherein L is the Euclidean distance between two control points, alpha is the included angle between the control points, and klimitIs the curvature limit.
5. The method according to claim 4, wherein in step 131),
the straight-going path guided by the driving behavior is composed of n1 control straight-line segments, and the straight-line segments meet the following constraints:
Figure FDA0002619243020000022
Figure FDA0002619243020000023
wherein lend-start1For Euclidean distance from the planning end point to the planning start point, li1Is the length of the i1 straight line segment, alphai1Is the included angle between the control points of the straight line segment of the i1 th segment;
the lane changing/merging road path determined according to the driving behavior is composed of n2 control straight line segments, and the straight line segments meet the following constraints:
Figure FDA0002619243020000031
Figure FDA0002619243020000032
wherein lend-start2For Euclidean distance from the planning end point to the planning start point, li2Is the length of the i2 straight line segment, alphai2Is the included angle between the control points of the straight line segment of the i2 th segment;
the turning path determined according to the driving behavior is composed of n3 control straight line segments, and the straight line segments meet the following constraints:
turning restraint without lane borrowing:
Figure FDA0002619243020000033
Figure FDA0002619243020000034
Figure FDA0002619243020000035
Figure FDA0002619243020000036
turning restraint of the left lane borrowing:
Figure FDA0002619243020000037
Figure FDA0002619243020000038
Figure FDA0002619243020000039
Figure FDA00026192430200000310
turning restraint of upward lane borrowing:
Figure FDA0002619243020000041
Figure FDA0002619243020000042
Figure FDA0002619243020000043
Figure FDA0002619243020000044
wherein, thetaj3Is the starting course angle, θg3Is the end point course angle, xg3、yg3As end point coordinate, xj3、yj3As coordinates of the starting point,/i3Is the length of the i3 straight line segment, alphai3Is the included angle between the control points of the straight line segment of the i3 th segment;
the turn-around path determined according to the driving behavior is composed of n4 control straight line segments, and the straight line segments meet the following constraints:
turning restraint without lane borrowing:
Figure FDA0002619243020000045
Figure FDA0002619243020000046
Figure FDA0002619243020000047
Figure FDA0002619243020000048
turning restraint of the left lane borrowing:
Figure FDA0002619243020000051
Figure FDA0002619243020000052
Figure FDA0002619243020000053
Figure FDA0002619243020000054
turning constraint of right lane borrowing:
Figure FDA0002619243020000055
Figure FDA0002619243020000056
Figure FDA0002619243020000057
Figure FDA0002619243020000058
turning restriction of lane borrowing on the left and the right:
Figure FDA0002619243020000059
Figure FDA00026192430200000510
Figure FDA00026192430200000511
Figure FDA00026192430200000512
wherein, thetaj4Is the starting course angle, θg4Is the end point course angle, xg4、yg4As end point coordinate, xj4、yj4As coordinates of the starting point,/i4Is the length of the i4 straight line segment, alphai4Is the included angle between the control points of the straight line segment of the i4 th segment.
6. The method as claimed in claim 4, wherein in the step 132), the turning and turning path according to the driving behavior modification is composed of an arc and two spiral lines, and the following constraints are satisfied:
the first helix:
a1s2+b1s+c1＝k1
c1＝kstart
Figure FDA0002619243020000061
Figure FDA0002619243020000062
Figure FDA0002619243020000063
the second helix:
a2s2+b2s+c2＝k2
c2＝kend
Figure FDA0002619243020000064
Figure FDA0002619243020000065
Figure FDA0002619243020000066
circular arc curve:
x1＝Rx-Rsinθ1
y＝Ry+Rcosθ1
x2＝Rx-Rsinθ2
y2＝Ry+Rsinθ2
where s is the path length, s1Is the length of the first helical line, s2Is the length of the second helical line, a1、b1、c1For the first spiral segment, a2、b2、c2For the second spiral, a characteristic parameter, k, is to be determined1Is the first section of the helix curvature, k2Is the curvature of the second spiral, kstartAs the curvature of origin, kendTo end point curvature, θiAs starting point heading, θendFor end course, θ1Is the end course of the first spiral line, theta2The starting course of the second spiral line, R is the radius of the circular arc, Rx，RyIs the center coordinate of the arc, x1，y1Coordinates of the start of the arc, x2，y2And (4) coordinates of the end point of the arc.
7. The fast and complete automatic driving trajectory planning method according to claim 3, wherein in the step 31), the path cutting specifically comprises:
and backtracking to a nearest non-collision control point along the basic path through the collision point detected by collision, wherein the non-collision control point is used as a starting point of the sampling/searching path plan, and a smooth path from the non-collision control point to the starting point of the basic path plan is used as an effective path obtained by cutting.
8. The method as claimed in claim 3, wherein in the step 3), the switching manner of the sampling/searching route planning includes user setting and adaptive selection according to environment complexity by the strategy of the route planning, when there are many obstacles in the environment or the environment map is large, the RRT is used for fast expansion, and when there are simple obstacles in the environment or the environment map is small, the a is used for heuristic search.
9. The method according to claim 8, wherein in step 34),
the sampling method is an improved RRT method, the node expansion modes of the method comprise three modes of forced expansion, driving behavior bias expansion and random expansion, and the specific self-adaptive selection methods of the three expansion modes are as follows:
when the environment is single and the success rate is higher than a set threshold value, adopting forced expansion; when the success rate of the multiple forced expansion is smaller than a set threshold value, adopting driving behavior bias expansion; when the success rate is low and the environment complexity is high, a standard expansion mode is adopted;
the searching method adopts an improved A method, the node expansion mode comprises three modes of driving behavior bias searching, steering angle constraint searching and standard searching, and the specific self-adaptive selection methods of the three expansion modes are as follows:
when the environment is single and the success rate is higher than a set threshold value, expanding the driving behavior tendency search; when the success rate of the driving behavior tendency search for many times is smaller than a set threshold value, adopting the expansion of steering angle constraint search; and when the success rate is low and the environment complexity is high, adopting an expansion mode of standard search.
10. The method as claimed in claim 1, wherein in step 4), the curvature extremum piecewise speed plan performs constant speed plan between two curvature extremum points of the basic path, and the acceleration and deceleration speed plan constrained by road adhesion and actuator performance is applied outside the extremum points,
or constant speed planning is carried out between the front point and the rear point of the four curvature extreme points of the basic path and the sampling/searching path, and acceleration and deceleration speed planning which is restricted by road surface adhesion and actuator performance is adopted outside the extreme points.
CN201811183196.7A 2018-10-11 2018-10-11 Rapid and complete automatic driving track planning method Active CN109540159B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201811183196.7A CN109540159B (en) 2018-10-11 2018-10-11 Rapid and complete automatic driving track planning method

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201811183196.7A CN109540159B (en) 2018-10-11 2018-10-11 Rapid and complete automatic driving track planning method

Publications (2)

Publication Number Publication Date
CN109540159A CN109540159A (en) 2019-03-29
CN109540159B true CN109540159B (en) 2020-11-27

Family

ID=65843846

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201811183196.7A Active CN109540159B (en) 2018-10-11 2018-10-11 Rapid and complete automatic driving track planning method

Country Status (1)

Country Link
CN (1) CN109540159B (en)

Families Citing this family (24)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109835336B (en) * 2019-02-19 2021-05-14 北京工业大学 Fuzzy algorithm-based wavy vehicle speed planning method with constraint square
CN110146883B (en) * 2019-05-17 2022-04-05 华东师范大学 Simultaneous and asynchronous multi-satellite platform MT-InSAR three-dimensional deformation decomposition method based on minimum acceleration
CN110187706A (en) * 2019-05-28 2019-08-30 上海钛米机器人科技有限公司 A kind of speed planning method, apparatus, electronic equipment and storage medium
CN110298131B (en) * 2019-07-05 2021-07-13 西南交通大学 Method for establishing automatic driving lane change decision model in hybrid driving environment
CN112506176B (en) * 2019-08-26 2024-05-28 上海汽车集团股份有限公司 Path planning method and device
CN110530390A (en) * 2019-09-16 2019-12-03 哈尔滨工程大学 A kind of non-particle vehicle path planning method under narrow environment
CN110716563A (en) * 2019-09-26 2020-01-21 山东理工大学 Electronic map given trajectory-based electric wheelchair path tracking control method and device
CN110703754B (en) * 2019-10-17 2021-07-09 南京航空航天大学 Path and speed highly-coupled trajectory planning method for automatic driving vehicle
CN110861650B (en) * 2019-11-21 2021-04-16 驭势科技（北京）有限公司 Vehicle path planning method and device, vehicle-mounted equipment and storage medium
CN113286983A (en) * 2019-12-20 2021-08-20 百度时代网络技术（北京）有限公司 Reference line smoothing method based on spline curve and spiral curve
CN111497827B (en) * 2020-02-17 2021-07-27 湖北亿咖通科技有限公司 Automatic parking method, device, medium and equipment
US11181917B2 (en) * 2020-04-15 2021-11-23 Baidu Usa Llc Path planning system based on steering wheel self zeroing for autonomous vehicles
CN111645677B (en) * 2020-05-20 2022-09-23 吉林大学 Vehicle braking and steering coordinated control emergency anti-collision system and control method
CN111504340B (en) * 2020-05-22 2022-04-29 北京汽车研究总院有限公司 Vehicle path planning method and device and vehicle
CN111750859B (en) * 2020-05-29 2021-11-05 广州极飞科技股份有限公司 Transition path planning method and related device
CN111879307A (en) * 2020-06-22 2020-11-03 国网河北省电力有限公司信息通信分公司 Vehicle path planning method based on vehicle body parameters and engineering construction information
CN112461240A (en) * 2020-11-11 2021-03-09 武汉理工大学 Unmanned aerial vehicle obstacle avoidance path planning method and system
CN112462785B (en) * 2020-12-04 2022-06-03 厦门大学 Mobile robot path planning method and device and storage medium
CN112632734A (en) * 2020-12-29 2021-04-09 武汉中海庭数据技术有限公司 High-precision map road connectivity test method and system
CN112379679B (en) * 2021-01-15 2021-04-23 北京理工大学 Unmanned vehicle local path planning method
CN112945254B (en) * 2021-01-21 2022-08-02 西北工业大学 Unmanned vehicle curvature continuous path planning method based on rapid expansion random tree
CN113320547B (en) * 2021-07-15 2023-08-25 广州小鹏自动驾驶科技有限公司 Path detection method and device and automobile
CN114089742B (en) * 2021-10-25 2023-09-26 广东嘉腾机器人自动化有限公司 AGV running speed control method, device and medium based on path curvature
CN117533354B (en) * 2023-12-28 2024-04-02 安徽蔚来智驾科技有限公司 Track generation method, driving control method, storage medium and intelligent device

Family Cites Families (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US9457807B2 (en) * 2014-06-05 2016-10-04 GM Global Technology Operations LLC Unified motion planning algorithm for autonomous driving vehicle in obstacle avoidance maneuver
CN104615889B (en) * 2015-02-09 2017-12-26 武汉大学 The intelligent vehicle path following method and system followed based on clothoid
FR3048517B1 (en) * 2016-03-07 2022-07-22 Effidence MOTORIZED AUTONOMOUS ROBOT WITH OBSTACLE ANTICIPATION
CN107063280B (en) * 2017-03-24 2019-12-31 重庆邮电大学 Intelligent vehicle path planning system and method based on control sampling
CN106873600A (en) * 2017-03-31 2017-06-20 深圳市靖洲科技有限公司 It is a kind of towards the local obstacle-avoiding route planning method without person bicycle
CN108229730B (en) * 2017-12-19 2021-07-20 同济大学 Unmanned vehicle track generation method based on fuzzy reward
CN108196536B (en) * 2017-12-21 2021-07-20 同济大学 Improved method for planning random tree path by fast searching through unmanned vehicle

Also Published As

Publication number Publication date
CN109540159A (en) 2019-03-29

Similar Documents

Publication Publication Date Title
CN109540159B (en) Rapid and complete automatic driving track planning method
CN108519773B (en) Path planning method for unmanned vehicle in structured environment
CN114234998B (en) Unmanned multi-target point track parallel planning method based on semantic road map
CN110298122B (en) Unmanned vehicle urban intersection left-turn decision-making method based on conflict resolution
CN107702716B (en) Unmanned driving path planning method, system and device
US11077878B2 (en) Dynamic lane biasing
CN107792065B (en) Method for planning road vehicle track
US11794736B2 (en) Dynamic collision checking
CN110304074B (en) Hybrid driving method based on layered state machine
CN108256233A (en) Intelligent vehicle trajectory planning and tracking and system based on driver style
EP4034440A1 (en) Blocking object avoidance
CN110333659B (en) Unmanned vehicle local path planning method based on improved A star search
CN110345957A (en) Vehicle route identification
CN109579854B (en) Unmanned vehicle obstacle avoidance method based on fast expansion random tree
JP2022506475A (en) Orbit generation
CN111152784B (en) Intelligent passenger-riding parking local path planning method
CN113608531B (en) Unmanned vehicle real-time global path planning method based on safety A-guidance points
CN114281084B (en) Intelligent vehicle global path planning method based on improved A-algorithm
CN113721637A (en) Intelligent vehicle dynamic obstacle avoidance path continuous planning method and system and storage medium
CN111397622A (en) Intelligent automobile local path planning method based on improved A-algorithm and Morphin algorithm
CN114898564A (en) Intersection multi-vehicle cooperative passing method and system under unstructured scene
CN116499486A (en) Complex off-road environment path planning method and system and electronic equipment
CN115077553A (en) Method, system, automobile, equipment and medium for planning track based on grid search
CN115140096A (en) Spline curve and polynomial curve-based automatic driving track planning method
CN117170377A (en) Automatic driving method and device and vehicle

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant