CN112238856B - Intelligent vehicle overtaking track optimization method based on hybrid particle swarm optimization - Google Patents

Intelligent vehicle overtaking track optimization method based on hybrid particle swarm optimization Download PDF

Info

Publication number
CN112238856B
CN112238856B CN202011092410.5A CN202011092410A CN112238856B CN 112238856 B CN112238856 B CN 112238856B CN 202011092410 A CN202011092410 A CN 202011092410A CN 112238856 B CN112238856 B CN 112238856B
Authority
CN
China
Prior art keywords
vehicle
overtaking
particle
track
time
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
CN202011092410.5A
Other languages
Chinese (zh)
Other versions
CN112238856A (en
Inventor
任师通
魏民祥
赵炳振
杨嘉伟
季桢杰
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.)
Nanjing University of Aeronautics and Astronautics
Original Assignee
Nanjing University of Aeronautics and Astronautics
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 Nanjing University of Aeronautics and Astronautics filed Critical Nanjing University of Aeronautics and Astronautics
Priority to CN202011092410.5A priority Critical patent/CN112238856B/en
Publication of CN112238856A publication Critical patent/CN112238856A/en
Application granted granted Critical
Publication of CN112238856B publication Critical patent/CN112238856B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B60—VEHICLES IN GENERAL
    • B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
    • B60W30/00—Purposes of road vehicle drive control systems not related to the control of a particular sub-unit, e.g. of systems using conjoint control of vehicle sub-units
    • B60W30/18—Propelling the vehicle
    • B60W30/18009—Propelling the vehicle related to particular drive situations
    • B60W30/18163—Lane change; Overtaking manoeuvres
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B60—VEHICLES IN GENERAL
    • B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
    • B60W50/00—Details of control systems for road vehicle drive control not related to the control of a particular sub-unit, e.g. process diagnostic or vehicle driver interfaces
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B60—VEHICLES IN GENERAL
    • B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
    • B60W50/00—Details of control systems for road vehicle drive control not related to the control of a particular sub-unit, e.g. process diagnostic or vehicle driver interfaces
    • B60W2050/0001—Details of the control system
    • B60W2050/0019—Control system elements or transfer functions
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B60—VEHICLES IN GENERAL
    • B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
    • B60W50/00—Details of control systems for road vehicle drive control not related to the control of a particular sub-unit, e.g. process diagnostic or vehicle driver interfaces
    • B60W2050/0001—Details of the control system
    • B60W2050/0019—Control system elements or transfer functions
    • B60W2050/0028—Mathematical models, e.g. for simulation
    • B60W2050/0031—Mathematical model of the vehicle

Landscapes

  • Engineering & Computer Science (AREA)
  • Automation & Control Theory (AREA)
  • Transportation (AREA)
  • Mechanical Engineering (AREA)
  • Human Computer Interaction (AREA)
  • Control Of Driving Devices And Active Controlling Of Vehicle (AREA)
  • Traffic Control Systems (AREA)

Abstract

The invention discloses an intelligent vehicle overtaking track optimization method based on a hybrid particle swarm algorithm, which comprises a vehicle environment sensing unit (a camera and a laser radar), a vehicle sensor unit (a speed sensor, an acceleration sensor and a front wheel steering angle sensor) and a vehicle electronic control unit; the method comprises the steps of firstly judging the feasibility of overtaking through a vehicle electronic control unit, further establishing a quintic polynomial overtaking track function, secondly establishing a target function based on the driving efficiency and stability of the vehicle, and finally optimizing the target function by adopting a hybrid particle swarm optimization to obtain the optimal overtaking driving track. The intelligent vehicle overtaking track optimization method based on the hybrid particle swarm optimization can enable the vehicle to quickly and stably complete overtaking tasks; the hybrid particle swarm algorithm can efficiently and accurately optimize the target function, and compared with other algorithms, the hybrid particle swarm algorithm is strong in convergence and short in optimization time.

Description

Intelligent vehicle overtaking track optimization method based on hybrid particle swarm optimization
Technical Field
The invention relates to the technical field of intelligent driving, in particular to a method for optimizing an intelligent vehicle overtaking track based on a hybrid particle swarm optimization.
Background
In recent years, intelligent vehicles have become one of the most concerned hot spots in the automotive industry due to the advantages of improving traffic efficiency, reducing traffic accidents and enhancing vehicle safety. The research on the intelligent vehicle mainly comprises three modules, namely a perception layer, a decision planning layer and a control layer. The overtaking is an important part in the driving process of the automobile, and according to incomplete statistics, the traffic accidents caused by overtaking each year account for more than 15 percent of all traffic accidents. In general, a driver performs overtaking behaviors depending on his own driving experience and the situation of the surrounding environment (vehicle speed, distance). However, for some drivers, it may make a wrong judgment on the feasibility of passing, thereby causing an accident to occur. The overtaking track planning is used as a key for safely and efficiently completing the overtaking task of the intelligent vehicle, so that the research on the overtaking track planning of the intelligent vehicle is of great significance for improving the vehicle passing efficiency.
Disclosure of Invention
The purpose of the invention is as follows: the invention aims to solve the technical problem of giving consideration to the defects of efficiency and safety in the overtaking process of an intelligent vehicle, and provides an overtaking track planning method optimized based on a hybrid particle swarm optimization algorithm.
The technical scheme is as follows:
an intelligent vehicle overtaking track optimization method based on a hybrid particle swarm algorithm comprises the following steps:
step 1: information acquisition, namely determining a geodetic coordinate system and a vehicle coordinate system according to road information acquired by a camera, wherein the road information comprises obstacle position information, lane speed limit information, vehicle information of a front vehicle and a self vehicle acquired by a laser radar, speed and acceleration information of the front vehicle and the self vehicle acquired by the laser radar and vehicle state information acquired by a vehicle sensor unit, and establishing an overtaking safe distance model;
step 2: the decision of overtaking is judged by the vehicle electronic control unit through judging the actual distance S between the self vehicle and the front vehiclerealSafe distance S from overtakingfThe difference of (a) is recorded as a; the camera detects the road information to identify the limited speed v of the lanel(ii) a Whether an obstacle exists in the left lane is detected through the camera, when the actual distance between the self-vehicle and the front vehicle meets the overtaking safety distance, the speeds of the two vehicles are both within the limited speed of the lanes, and the left overtaking lane has no obstacle, the vehicle electronic control unit sends a signal to the vehicle execution mechanism to start overtaking; the vehicle electronic control unit sends the overtaking signal to a vehicle actuating mechanism;
and step 3: establishing overtaking track function, establishing quintic polynomial overtaking track function, determining initial time state
Figure BDA0002722586030000021
And end time status
Figure BDA0002722586030000022
Substituting the numerical values into a quintic polynomial overtaking track function to obtain a function formula containing three parameter variables of initial vehicle speed, lane change longitudinal track length and lane change time;
and 4, step 4: and (3) optimizing the parameter variables, namely performing parameter optimization on the two parameter variables of the lane change longitudinal track length and the lane change required time by using a hybrid particle swarm algorithm to obtain the optimal overtaking tracks at different initial speeds.
Further, in the step 1, the camera transmits the collected road information, the relative position between the front vehicle and the self vehicle, the speed of the front vehicle and the acceleration information collected by the laser radar to the vehicle electronic control unit, the vehicle sensor comprises a vehicle speed sensor, a front wheel steering angle sensor and an acceleration sensor which respectively collect the speed, the front wheel steering angle and the acceleration of the vehicle, and the obtained information is transmitted to the vehicle electronic control unit;
further, the overtaking safe distance model in the step 2 is as follows:
Figure BDA0002722586030000023
in the formula, vfIs the speed of the bicycle, vrIs the relative speed between the vehicle and the front vehicle, mu is the road adhesion coefficient, g is the gravity acceleration, t1Delay time for driver reaction, t2For the brake on time, d is the minimum stopping distance.
Further, the concrete implementation of step 3 includes dividing the overtaking track function into three stage functions of lane change, overtaking and lane merging, considering the lane change stage track and the lane merging stage track as symmetrical in order to reduce the calculation amount, and the overtaking stage track is a straight line without optimization, so that the whole overtaking track can be obtained only by optimizing the lane change stage, and establishing a quintic polynomial lane change track function:
Figure BDA0002722586030000024
in the formula, a0,a1,a2,a3,a4,a5,b0,b1,b2,b3,b4,b5Each term representing the above polynomial functionCoefficient of determining the initial time t0The state is as follows:
Figure BDA0002722586030000025
end time tzThe state is as follows:
Figure BDA0002722586030000026
the coefficients of the polynomial may be expressed as:
a0＝0,a1＝v0,a2＝0,a3＝10(xz-v0tz)/tz 3,a4＝-15(xz-v0tz)/tz 4,a5＝6(xz-v0tz)/tz 5
b0＝0,b1＝0,b2＝0,b3＝10yz/tz 3,b4＝-15yz/tz 4,b5＝yz/tz 5
in the formula, v0Longitudinal speed, x, representing the initial time of the vehiclezIndicating the length of the longitudinal track, t, of the lane changezIndicates the time required for lane change, yzIndicating the width of the lane, typically 3.75m
Then the lane change curve function can be found as:
Figure BDA0002722586030000031
because the third-stage function and the first-stage function in the overtaking track are symmetrical, the overtaking track function can be obtained as follows:
Figure BDA0002722586030000032
in the formula, t1,t2,t3Respectively showing the finish time of the first stage, the second stage and the third stage of overtaking.
Further, the specific implementation of step 4 includes:
step 4.1, population initialization and individual coding
Initializing a population with n number of particles in a certain search space, wherein the ith particle represents an m-dimensional vector
Figure BDA0002722586030000034
i is 1,2, …, n, the positions of the particles are all solutions of the equation, the speed of the ith particle moving in the search space is also an m-dimensional vector, which is
Figure BDA0002722586030000035
The method is characterized in that a coding mode is adopted for particle individuals in a population, and each particle comprises two parameter variables, namely channel changing longitudinal track length and channel changing total time.
Step 4.2, individual fitness value calculation
The calculation formula of the individual particle fitness value is shown as follows, wherein the fitness value is defined as the following formula (f),
j denotes an objective function, which is used to determine the parameter variables longitudinal track length, total track change time, ayThe second derivative can be obtained by directly solving the following formula,
j represents an objective function, the longitudinal track length and the total lane change time of the parameter variables are determined by utilizing the objective function, the objective function is established based on longitudinal displacement, lateral acceleration, yaw velocity, mass center and lateral deviation angle and front wheel steering angle in consideration of the stability problem of the automobile, the five indexes can be obtained by a fifth-order polynomial function and only comprise two variables of the longitudinal track length and the total lane change time,
Figure BDA0002722586030000033
ωrcan be obtained by the following formula,
Figure BDA0002722586030000041
wherein v isxIs the vehicle longitudinal speed, R is the vehicle turning radius;
beta can be obtained by the following formula,
Figure BDA0002722586030000042
the delta can be obtained by the following formula,
Figure BDA0002722586030000043
wherein
Figure BDA0002722586030000044
k1,k2The cornering stiffness of the front and rear wheels is indicated, a, b indicate the distance of the centre of mass to the front and rear axles respectively, L indicates the wheelbase, m indicates the total mass of the vehicle, i is the vehicle steering ratio, and i is typically taken to be 17.
An objective function expression can thus be established:
Figure BDA0002722586030000045
in the formula, xzIndicating the length of the longitudinal track of the lane change, ayIndicating lateral acceleration, omega, at any time of lane changerRepresenting the yaw velocity at any time of changing the track, beta representing the centroid slip angle at any time of changing the track, delta representing the front wheel corner at any time of changing the track, and xamax、aymax、ωrmax、βmax、δmaxRespectively represents the maximum longitudinal track length, the maximum lateral acceleration, the maximum yaw rate, the maximum mass center slip angle and the maximum front wheel rotation angle, omega, allowed in all the lane changing tracks1、ω2、ω3、ω4、ω5Is a weight coefficient, and ω1+ω2+ω3+ω4+ω5＝1；
I.e. the fitness value of the particle is the minimum of the objective function, will
Figure BDA0002722586030000046
Substituting the objective function to obtain a fitness value corresponding to the particle, and judging the quality of the solution according to the obtained fitness value;
step 4.3, searching individual extremum and group extremum
By calculating the fitness value of each particle and the positions the particles have undergone, the individual extremum and the group extremum of the particles can be obtained, and the optimal position searched by the ith particle is made to be
Figure BDA0002722586030000047
The optimum position searched in the global is
Figure BDA0002722586030000048
Passing the current fitness value of the particle and the particle through the optimal position
Figure BDA0002722586030000049
The fitness value of the particle is compared, the current fitness value of the particle is selected as the current optimal position of the particle according to the comparison result, meanwhile, the current fitness value of the particle is compared, the current fitness value of the particle is selected as the global optimal position of the particle according to the comparison result, and the method can be determined by the following formula:
Figure BDA0002722586030000051
wherein, Xi(t +1) denotes an s-dimensional vector at the moment of the ith particle t +1, Pi(t) represents an optimal position of the ith particle in the search space;
step 4.4, update the speed and position of the particle
The particle is updated by the following formula:
vim(t+1)＝vim(t+1)+c1r1m(t)(pim(t)-xim(t))+c2r2m(t)(pgm(t)-xgm(t))
xim(t+1)＝xim(t)+vim(t+1)
in the formula: v. ofim(t +1) represents the velocity of the ith particle at time t +1, xim(t) represents the position of the ith particle at time t +1, i ═ 1, n],s＝[1,m]Learning factor c1And c2Is a non-negative constant; r is1And r2Obey [0,1 ] for mutually independent pseudo-random numbers]V is uniformly distributed overim∈[-vmax,vmax]Wherein v ismaxIs a constant set by the user;
step 4.5, crossover operation
Selecting two crossing positions, crossing the individual and the individual extreme value or the individual and the group extreme value, adopting the principle of reserving excellence for the obtained new particle, and only updating when the fitness value of the new particle is superior to that of the old particle;
step 4.6, mutation operation
The two positions of variation st1 and st2 within the individual particles are chosen, and the two positions are interchanged, again taking the principle of preserving excellence for the new particles obtained, and are only updatable when the fitness value of the new particle is better than that of the old one.
Determining an initial state and an objective function, considering the problem of automobile stability, establishing the objective function based on longitudinal displacement, lateral acceleration, yaw velocity, mass center lateral deviation angle and front wheel steering angle, optimizing two parameter variables of lane changing longitudinal track length and lane changing time of the objective function through a hybrid particle swarm optimization, and finally obtaining an optimal overtaking track;
and (4) solving the lane change longitudinal track length and the lane change required time under different vehicle speeds through the optimization characteristic in the hybrid particle swarm optimization, and further obtaining the optimal overtaking track function.
Has the advantages that:
1. the track optimization algorithm provided by the invention gives consideration to the principles of driving efficiency, vehicle stability, comfort and the like, so that the vehicle can quickly and stably complete overtaking tasks;
2. the adopted hybrid particle swarm algorithm can efficiently and accurately optimize the target function, and compared with other algorithms, the hybrid particle swarm algorithm has strong convergence and shorter optimization time;
3. and the three running conditions of low speed, medium speed and high speed are analyzed respectively, so that the model can be suitable for more complex conditions.
Drawings
FIG. 1 is a logic diagram of vehicle cut-in trajectory optimization;
FIG. 2 is a schematic illustration of a vehicle cut-in behavior;
FIG. 3 is a graph of the results of a hybrid particle swarm optimization;
FIG. 4 is a graph of optimal cut-in trajectories at different longitudinal vehicle speeds.
Detailed Description
The following detailed description of embodiments of the invention refers to the accompanying drawings.
As shown in fig. 1, fig. 1 is a logic block diagram of the system, and is characterized by comprising an environment sensing unit, a vehicle sensor unit, a vehicle electronic control unit, a trajectory planning module, and a hybrid particle swarm algorithm, wherein the objective function is optimized and solved to obtain an optimal overtaking trajectory.
Step 1: acquiring road information and vehicle state information according to the camera, the laser radar and the vehicle sensor unit, and determining a geodetic coordinate system and a vehicle coordinate system;
step 2: the vehicle electronic control unit processes the received camera information, the laser radar information and the information of each sensor of the vehicle, and establishes a overtaking safe distance model;
and step 3: the vehicle electronic control unit sends the overtaking signal to the vehicle executing mechanism by processing the information;
and 4, step 4: establishing a quintic polynomial overtaking track model, and optimizing the overtaking track through a hybrid particle swarm optimization to finally obtain an optimal overtaking track;
further, in the step 1, acquiring lane line information and lane limit speed information through a camera, transmitting the acquired information to a vehicle electronic control unit, acquiring front vehicle information, adjacent lane information, the relative position of a front vehicle and the front vehicle, the speed and acceleration information of the front vehicle by a laser radar, transmitting the acquired information to the vehicle electronic control unit, and acquiring the driving speed, the front wheel steering angle and the acceleration of the vehicle by a vehicle speed sensor, a front wheel steering angle sensor and an acceleration sensor in the vehicle;
further, in step 2, the camera information, the laser radar information and the information of each sensor of the vehicle received by the vehicle electronic control unit are processed to establish a overtaking safe distance model;
Figure BDA0002722586030000061
in the formula, vfIs the speed of the bicycle, vrIs the relative speed between the vehicle and the front vehicle, mu is the road adhesion coefficient, g is the gravity acceleration, t1Delay time for driver reaction, t2For the brake on time, d is the minimum stopping distance.
Further, in step 3, the vehicle electronic control unit judges the actual distance S between the own vehicle and the front vehiclerealSafe distance S from overtakingfThe difference of (a) is denoted as a,
further, the road information is detected through a camera, and the lane limit speed v is recognizedl，
Furthermore, whether an obstacle exists in the left lane or not is detected through a camera,
when the actual distance between the self vehicle and the front vehicle meets the overtaking safety distance, the speeds of the two vehicles are both within the limited speed of the lane, and the left overtaking lane has no obstacle, the vehicle electronic control unit sends a signal to the vehicle executing mechanism to start overtaking;
further, in step 4, as shown in FIG. 2, the center line of the passing lane is entered from the center line of the passing lane (first passing stage: lane changing stage), and after the passing lane passes the preceding vehicle (second passing stage: passing stage), the passing lane is returned to the original lane (third passing stage: merging stage). Therefore, the overtaking behaviors can be regarded as two-time lane changing behaviors and one-time overtaking behaviors, in order to reduce the calculation amount, the two-time lane changing behaviors are regarded as symmetrical, the overtaking stage is a straight line, optimization is not needed, namely, only one-time optimization is carried out on the lane changing track in the first overtaking stage (lane changing stage), and the optimal overtaking track can be obtained.
Further, in step 4, as shown in fig. 2, the vehicle lane is set to be the abscissa of the geodetic coordinate system, the ordinate is perpendicular to the vehicle body, the initial position of the center of mass of the vehicle is set to be the origin, the path formula of the vehicle lane is y-0, and the path formula of the passing lane is y-yz，yzIs the width of a lane and is generally taken as (y)z＝3.75m)。
Further, in step 4, a fifth-order polynomial function is used as a lane change curve, that is, a lane change equation is as follows:
Figure BDA0002722586030000071
determining an initial time t0Comprises the following steps:
Figure BDA0002722586030000072
end time tzComprises the following steps:
Figure BDA0002722586030000073
the coefficients of the polynomial may be expressed as:
Figure BDA0002722586030000074
then the lane change curve function can be found as:
Figure BDA0002722586030000081
in summary, the overtaking trajectory function can be obtained as follows:
Figure BDA0002722586030000082
in the formula, t1,t2,t3Respectively showing the finish time of the first stage, the second stage and the third stage of overtaking.
For the first phase of the overtaking process, the longitudinal speed v is due to0Longitudinal track length xzAnd total time t required for lane changezNumerous lane change curves can be obtained. Therefore, the parameters of the lane-changing function need to be optimized to obtain the optimal lane-changing function curve.
Further, in step 4, an objective function is set:
Figure BDA0002722586030000083
in the formula, xzIndicating the length of the longitudinal track of the lane change, ayIndicating lateral acceleration, omega, at any time of lane changerRepresenting the yaw velocity at any time of changing the track, beta representing the centroid slip angle at any time of changing the track, delta representing the front wheel corner at any time of changing the track, and xamax、aymax、ωrmax、βmax、δmaxRespectively represents the maximum longitudinal track length, the maximum lateral acceleration, the maximum yaw rate, the maximum mass center slip angle and the maximum front wheel rotation angle, omega, allowed in all the lane changing tracks1、ω2、ω3、ω4、ω5Is a weight coefficient, and ω1+ω2+ω3+ω4+ω5＝1
Further, a in the above objective functionyThe second derivative can be obtained by directly solving the following formula,
Figure BDA0002722586030000084
further, ω in the above objective functionrCan be obtained by the following formula,
Figure BDA0002722586030000085
further, β in the above objective function can be obtained by the following formula,
Figure BDA0002722586030000086
further, δ in the above objective function can be obtained by the following formula,
Figure BDA0002722586030000091
wherein
Figure BDA0002722586030000092
k1,k2The cornering stiffness of the front and rear wheels is indicated, a, b indicate the distance of the centre of mass to the front and rear axles respectively, m indicates the total mass of the vehicle, and i is the vehicle steering ratio, typically taken as i-17.
Further, in step 4, the vehicle needs to meet the stability requirement during the driving process, so the constraint conditions are set as follows:
Figure BDA0002722586030000093
further, in step 4, the objective function is optimized through the hybrid particle swarm algorithm, a hybrid particle swarm algorithm combining a Genetic Algorithm (GA) and a standard particle swarm algorithm (PSO) is provided, a method that the standard particle swarm algorithm updates the positions of the particles through tracking extreme values is abandoned, and the optimal solution is searched in a mode that the particles are crossed with the individual extreme values and the group extreme values and the particles are mutated by adding crossing and mutation operations in the genetic algorithm.
Further, the concrete steps of solving by the hybrid particle swarm algorithm are as follows:
step 4.1, population initialization and individual coding
Initializing a population with n number of particles in a certain search space, wherein the ith particle represents an m-dimensional vector
Figure BDA0002722586030000094
The position of the particle is the solution of the equation. The velocity of the ith particle moving in the search space is also an m-dimensional vector, of
Figure BDA0002722586030000095
And (3) adopting a coding mode for the particle individuals in the population, wherein each particle comprises two independent variables of the lane change longitudinal track length and the lane change total time.
Step 4.2, individual fitness value calculation
The calculation formula of the fitness value of the particle individual is shown as follows, wherein the fitness value of the particle individual is as follows
I.e. the fitness value of the particle is the minimum of the objective function. Will be provided with
Figure BDA0002722586030000096
The fitness value corresponding to the particle can be obtained by substituting the objective function, and the quality of the solution is determined according to the obtained fitness value.
Step 4.3, searching individual extremum and group extremum
By calculating the fitness value of each particle and the positions the particles have undergone, the individual extremum and the group extremum of the particles can be obtained, and the optimal position searched by the ith particle is made to be
Figure BDA0002722586030000101
The optimum position searched in the global is
Figure BDA0002722586030000102
Passing the current fitness value of the particle and the particle through the optimal position
Figure BDA0002722586030000103
And comparing the fitness values, and selecting the fitness values as the current optimal positions of the particles according to the comparison result. Meanwhile, the current fitness value of the particle is compared, and the current fitness value of the particle is compared according to the comparisonAs a result, it is selected as the global optimal position for the particle. This can be determined by the following equation:
Figure BDA0002722586030000104
step 4.4, update the speed and position of the particle
The particle is updated by the following formula:
vim(t+1)＝vim(t+1)+c1r1m(t)(pim(t)-xim(t))+c2r2m(t)(pgm(t)-xgm(t))
xim(t+1)＝xim(t)+vim(t+1)
in the formula: i ═ 1, n],s＝[1,m]Learning factor c1And c2Is a non-negative constant; r is1And r2Obey [0,1 ] for mutually independent pseudo-random numbers]Are uniformly distributed. v. ofim∈[-vmax,vmax]Wherein v ismaxIs a constant set by itself.
Step 4.5, crossover operation
Two crossing positions are selected, individuals and individual extrema or individuals and group extrema are crossed, an excellent principle is kept for the obtained new particles, and the new particles can be updated only when the fitness value of the new particles is superior to that of the old particles.
Step 4.6, mutation operation
The two positions of variation st1 and st2 within the individual particles are chosen, and the two positions are interchanged, again taking the principle of preserving excellence for the new particles obtained, and are only updatable when the fitness value of the new particle is better than that of the old one.
Further, through the optimization characteristic in the hybrid particle swarm optimization, the obtained relationship between the iteration times and the fitness value is shown in fig. 3, the obtained fitness is used for solving the lane change longitudinal track length and the lane change required time under different vehicle speeds, and the optimal overtaking track function is obtained as shown in the following table.
Further, three typical vehicle speeds of high speed, medium speed and low speed are selected for research, and the obtained overtaking track curve is shown in fig. 3.
TABLE 1 track optimization results
Figure BDA0002722586030000105
Figure BDA0002722586030000111
The foregoing is only a preferred embodiment of the present invention, and it should be noted that, for those skilled in the art, various modifications and decorations can be made without departing from the principle of the present invention, and these modifications and decorations should also be regarded as the protection scope of the present invention.

Claims (5)

1. An intelligent vehicle overtaking track optimization method based on a hybrid particle swarm optimization algorithm is characterized by comprising the following steps:
step 1: information acquisition, namely determining a geodetic coordinate system and a vehicle coordinate system according to road information acquired by a camera, wherein the road information comprises obstacle position information, lane speed limit information, vehicle information of a front vehicle and a self vehicle acquired by a laser radar and vehicle state information acquired by a vehicle sensor unit, and establishing an overtaking safe distance model;
step 2: the decision of overtaking is judged by the vehicle electronic control unit through judging the actual distance S between the self vehicle and the front vehiclerealSafe distance S from overtakingfThe difference of (a) is recorded as a; the camera detects the road information to identify the limited speed v of the lanel(ii) a Whether an obstacle exists in the left lane is detected through the camera, when the actual distance between the self-vehicle and the front vehicle meets the overtaking safety distance, the speeds of the two vehicles are both within the limited speed of the lanes, and the left overtaking lane has no obstacle, the vehicle electronic control unit sends a signal to the vehicle execution mechanism to start overtaking; the vehicle electronic control unit sends the overtaking signal to a vehicle actuating mechanism;
and step 3: establishing overtaking track function, establishing quintic polynomial overtaking track function, determining initial time state
Figure FDA0003409657190000011
And end time status
Figure FDA0003409657190000012
Substituting the numerical values into a quintic polynomial overtaking track function to obtain a function formula containing three parameter variables of initial vehicle speed, lane change longitudinal track length and lane change time;
and 4, step 4: and (3) optimizing the parameter variables, namely performing parameter optimization on the two parameter variables of the lane change longitudinal track length and the lane change required time by using a hybrid particle swarm algorithm to obtain the optimal overtaking tracks at different initial speeds.
2. The hybrid particle swarm optimization-based intelligent vehicle overtaking track optimization method according to claim 1, wherein the camera in step 1 transmits the collected road information and the collected relative position between the front vehicle and the self vehicle, the collected speed and acceleration information of the front vehicle to the vehicle electronic control unit through the laser radar, the vehicle sensor comprises a vehicle speed sensor, a front wheel rotation angle sensor and an acceleration sensor which respectively collect the speed, the front wheel rotation angle and the acceleration of the vehicle, and the obtained information is transmitted to the vehicle electronic control unit.
3. The hybrid particle swarm optimization-based intelligent vehicle overtaking track optimization method according to claim 1, wherein the overtaking safe distance model in the step 2 is as follows:
Figure FDA0003409657190000013
in the formula, vfIs the speed of the bicycle, vrIs the relative speed between the vehicle and the front vehicle, mu is the road adhesion coefficient, g is the gravityAcceleration, t1Delay time for driver reaction, t2For the brake on time, d is the minimum stopping distance.
4. The intelligent vehicle overtaking track optimization method based on the hybrid particle swarm optimization according to claim 1 is characterized in that the specific implementation of step 3 includes that an overtaking track function is divided into three stage functions of lane change, overtaking and lane merging, in order to reduce the calculation amount, the lane change stage track and the lane merging stage track are regarded as symmetrical, and the overtaking stage track is a straight line and does not need to be optimized, so that the whole overtaking track can be obtained by optimizing only a lane change stage, and a quintic polynomial lane change track function is established:
Figure FDA0003409657190000021
in the formula, a0,a1,a2,a3,a4,a5,b0,b1,b2,b3,b4,b5Determining initial time t by representing each coefficient of the polynomial function0The state is as follows:
Figure FDA0003409657190000022
end time tzThe state is as follows:
Figure FDA0003409657190000023
the coefficients of the polynomial may be expressed as:
a0＝0,a1＝v0,a2＝0,a3＝10(xz-v0tz)/tz 3,a4＝-15(xz-v0tz)/tz 4,a5＝6(xz-v0tz)/tz 5
b0＝0,b1＝0,b2＝0,b3＝10yz/tz 3,b4＝-15yz/tz 4,b5＝yz/tz 5
in the formula, v0Longitudinal speed, x, representing the initial time of the vehiclezIndicating the length of the longitudinal track, t, of the lane changezIndicates the time required for lane change, yzRepresenting the lane width, taking 3.75m, the lane change curve function can be found as:
Figure FDA0003409657190000024
because the third-stage function and the first-stage function in the overtaking track are symmetrical, the overtaking track function can be obtained as follows:
Figure FDA0003409657190000025
Figure FDA0003409657190000031
in the formula, t1,t2,t3Respectively showing the finish time of the first stage, the second stage and the third stage of overtaking.
5. The intelligent vehicle overtaking track optimization method based on the hybrid particle swarm optimization according to claim 1, wherein the concrete implementation of the step 4 comprises the following steps:
step 4.1, population initialization and individual coding
Initializing a population with n number of particles in a certain search space, wherein the ith particle represents an m-dimensional vector
Figure FDA0003409657190000032
The positions of the particles are all solutions of equations, and the ith particle moves in the search spaceThe velocity of motion is also a vector of m dimensions, of
Figure FDA0003409657190000033
The method comprises the steps that a coding mode is adopted for particle individuals in a population, and each particle comprises two parameter variables, namely channel changing longitudinal track length and channel changing total time;
step 4.2, individual fitness value calculation
The calculation formula of the individual particle fitness value is shown as follows, wherein the fitness value is defined as the following formula (f),
j represents an objective function, which is used to determine the parameter variables, longitudinal track length, total time to switch tracks,
aythe second derivative can be obtained by directly solving the following formula,
Figure FDA0003409657190000034
ωrcan be obtained by the following formula,
Figure FDA0003409657190000035
wherein v isxIs the vehicle longitudinal speed, R is the vehicle turning radius;
beta can be obtained by the following formula,
Figure FDA0003409657190000036
the delta can be obtained by the following formula,
Figure FDA0003409657190000037
wherein
Figure FDA0003409657190000041
k1,k2Representing the cornering stiffness of the front wheel and the rear wheel, a and b respectively representing the distances from the center of mass to the front axle and the rear axle, L representing the wheelbase, m representing the total mass of the automobile, i representing the transmission ratio of the steering system of the automobile, and i being 17;
an objective function expression can thus be established:
Figure FDA0003409657190000042
in the formula, xzIndicating the length of the longitudinal track of the lane change, ayIndicating lateral acceleration, omega, at any time of lane changerRepresenting the yaw velocity at any time of changing the track, beta representing the centroid slip angle at any time of changing the track, delta representing the front wheel corner at any time of changing the track, and xamax、aymax、ωrmax、βmax、δmaxRespectively represents the maximum longitudinal track length, the maximum lateral acceleration, the maximum yaw rate, the maximum mass center slip angle and the maximum front wheel rotation angle, omega, allowed in all the lane changing tracks1、ω2、ω3、ω4、ω5Is a weight coefficient, and ω1+ω2+ω3+ω4+ω5＝1；
I.e. the fitness value of the particle is the minimum of the objective function, will
Figure FDA0003409657190000047
Substituting the objective function to obtain a fitness value corresponding to the particle, and judging the quality of the solution according to the obtained fitness value;
step 4.3, searching individual extremum and group extremum
By calculating the fitness value of each particle and the positions the particles have undergone, the individual extremum and the group extremum of the particles can be obtained, and the optimal position searched by the ith particle is made to be
Figure FDA0003409657190000043
The optimum position searched in the global is
Figure FDA0003409657190000044
Passing the current fitness value of the particle and the particle through the optimal position
Figure FDA0003409657190000045
The fitness value of the particle is compared, the current fitness value of the particle is selected as the current optimal position of the particle according to the comparison result, meanwhile, the current fitness value of the particle is compared, the current fitness value of the particle is selected as the global optimal position of the particle according to the comparison result, and the method can be determined by the following formula:
Figure FDA0003409657190000046
wherein, Xi(t +1) denotes an s-dimensional vector at the moment of the ith particle t +1, Pi(t) represents an optimal position of the ith particle in the search space;
step 4.4, update the speed and position of the particle
The particle is updated by the following formula:
vim(t+1)＝vim(t+1)+c1r1m(t)(pim(t)-xim(t))+c2r2m(t)(pgm(t)-xgm(t))
xim(t+1)＝xim(t)+vim(t+1)
in the formula: v. ofim(t +1) represents the velocity of the ith particle at time t +1, xim(t) represents the position of the ith particle at time t +1, i ═ 1, n],s＝[1,m]Learning factor c1And c2Is a non-negative constant; r is1And r2Obey [0,1 ] for mutually independent pseudo-random numbers]V is uniformly distributed overim∈[-vmax,vmax]Wherein v ismaxIs a constant set by the user;
step 4.5, crossover operation
Selecting two crossing positions, crossing the individual and the individual extreme value or the individual and the group extreme value, adopting the principle of reserving excellence for the obtained new particle, and only updating when the fitness value of the new particle is superior to that of the old particle;
step 4.6, mutation operation
Selecting two variation positions st1 and st2 in the particle individuals, wherein the two selected positions are interchanged, and adopting the principle of reserving excellent property for obtaining new particles, and the new particles can be updated only when the adaptability value of the new particles is better than that of the old particles;
determining an initial state and an objective function, considering the problem of automobile stability, establishing the objective function based on longitudinal displacement, lateral acceleration, yaw velocity, mass center lateral deviation angle and front wheel steering angle, optimizing two parameter variables of lane changing longitudinal track length and lane changing time of the objective function through a hybrid particle swarm optimization, and finally obtaining an optimal overtaking track;
and (4) solving the lane change longitudinal track length and the lane change required time under different vehicle speeds through the optimization characteristic in the hybrid particle swarm optimization, and further obtaining the optimal overtaking track function.
CN202011092410.5A 2020-10-13 2020-10-13 Intelligent vehicle overtaking track optimization method based on hybrid particle swarm optimization Active CN112238856B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202011092410.5A CN112238856B (en) 2020-10-13 2020-10-13 Intelligent vehicle overtaking track optimization method based on hybrid particle swarm optimization

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202011092410.5A CN112238856B (en) 2020-10-13 2020-10-13 Intelligent vehicle overtaking track optimization method based on hybrid particle swarm optimization

Publications (2)

Publication Number Publication Date
CN112238856A CN112238856A (en) 2021-01-19
CN112238856B true CN112238856B (en) 2022-04-05

Family

ID=74169019

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202011092410.5A Active CN112238856B (en) 2020-10-13 2020-10-13 Intelligent vehicle overtaking track optimization method based on hybrid particle swarm optimization

Country Status (1)

Country Link
CN (1) CN112238856B (en)

Families Citing this family (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113674529A (en) * 2021-10-25 2021-11-19 北京理工大学 Autonomous overtaking method and system
CN114384912A (en) * 2022-01-12 2022-04-22 河南科技大学 Intelligent vehicle overtaking path planning method based on improved artificial potential field algorithm
CN114291092B (en) * 2022-01-26 2023-05-16 中国联合网络通信集团有限公司 Vehicle lane change control method and device, electronic control unit and storage medium

Family Cites Families (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102609765A (en) * 2012-03-22 2012-07-25 北京工业大学 Intelligent vehicle lane change path planning method based on polynomial and radial basis function (RBF) neural network
CN106671982B (en) * 2017-01-09 2019-05-17 厦门大学 Driverless electric automobile automatic overtaking system system and method based on multiple agent
CN109283843B (en) * 2018-10-12 2021-08-03 江苏大学 Path-changing trajectory planning method based on fusion of polynomial and particle swarm optimization
CN110614998B (en) * 2019-08-21 2021-01-08 南京航空航天大学 Aggressive driving-assisted curve obstacle avoidance and road changing path planning system and method

Also Published As

Publication number Publication date
CN112238856A (en) 2021-01-19

Similar Documents

Publication Publication Date Title
CN111717204B (en) Lateral control method and system for automatic driving vehicle
CN112238856B (en) Intelligent vehicle overtaking track optimization method based on hybrid particle swarm optimization
CN111717192B (en) Control method and system for automatically driving vehicle
CN109669461B (en) Decision-making system for automatically driving vehicle under complex working condition and track planning method thereof
CN110979305B (en) Vehicle abnormal lane change control method, device and system
CN110187639B (en) Trajectory planning control method based on parameter decision framework
CN108256233B (en) Intelligent vehicle trajectory planning and tracking method and system based on driver style
CN113276848B (en) Intelligent driving lane changing and obstacle avoiding track planning and tracking control method and system
CN110928297B (en) Intelligent bus route planning method based on multi-objective dynamic particle swarm optimization
Hima et al. Trajectory tracking for highly automated passenger vehicles
CN108717268A (en) Automatic Pilot minimum time maneuver control system and its control method based on optimum control and safe distance
CN110161865B (en) Intelligent vehicle track change trajectory planning method based on nonlinear model predictive control
CN112578672B (en) Unmanned vehicle trajectory control system based on chassis nonlinearity and trajectory control method thereof
CN115285145A (en) Unmanned curve collision avoidance trajectory planning and tracking control method
Zhang et al. Structured road-oriented motion planning and tracking framework for active collision avoidance of autonomous vehicles
CN114194215A (en) Intelligent vehicle obstacle avoidance and track changing planning method and system
CN116499486A (en) Complex off-road environment path planning method and system and electronic equipment
CN113848942A (en) Constraint-oriented intelligent network-connected automobile robust lane-changing confluence control method
Hao et al. The eco-driving considering coordinated control strategy for the intelligent electric vehicles
Kone Lateral and longitudinal control of an autonomous racing vehicle.
CN115447615A (en) Trajectory optimization method based on vehicle kinematics model predictive control
CN115042770A (en) Vehicle queue transverse control method based on distributed robust model prediction
Nossier et al. Multi-obstacle avoidance algorithm for autonomous vehicles
Jeong Path Tracking Control for Four-Wheel-Steering Autonomous Vehicles based on Adaptive Sliding Mode Control with Control Allocation
Liu et al. Optimal control of path tracking for vehicle-handling dynamics

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