CN107813820A - A kind of unmanned vehicle lane-change paths planning method for imitating outstanding driver - Google Patents
A kind of unmanned vehicle lane-change paths planning method for imitating outstanding driver Download PDFInfo
- Publication number
- CN107813820A CN107813820A CN201710953896.9A CN201710953896A CN107813820A CN 107813820 A CN107813820 A CN 107813820A CN 201710953896 A CN201710953896 A CN 201710953896A CN 107813820 A CN107813820 A CN 107813820A
- Authority
- CN
- China
- Prior art keywords
- path
- lane
- neural network
- vehicle
- driving
- 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.)
- Pending
Links
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, or advanced driver assistance systems for ensuring comfort, stability and safety or drive control systems for propelling or retarding the vehicle
- B60W30/08—Active safety systems predicting or avoiding probable or impending collision or attempting to minimise its consequences
- B60W30/09—Taking automatic action to avoid collision, e.g. braking and steering
-
- 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
- B60W10/00—Conjoint control of vehicle sub-units of different type or different function
- B60W10/18—Conjoint control of vehicle sub-units of different type or different function including control of braking systems
-
- 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
- B60W10/00—Conjoint control of vehicle sub-units of different type or different function
- B60W10/20—Conjoint control of vehicle sub-units of different type or different function including control of steering systems
-
- 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, or advanced driver assistance systems for ensuring comfort, stability and safety or drive control systems for propelling or retarding the vehicle
- B60W30/08—Active safety systems predicting or avoiding probable or impending collision or attempting to minimise its consequences
- B60W30/095—Predicting travel path or likelihood of collision
- B60W30/0953—Predicting travel path or likelihood of collision the prediction being responsive to vehicle dynamic parameters
-
- 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, or advanced driver assistance systems for ensuring comfort, stability and safety or drive control systems for propelling or retarding the vehicle
- B60W30/08—Active safety systems predicting or avoiding probable or impending collision or attempting to minimise its consequences
- B60W30/095—Predicting travel path or likelihood of collision
- B60W30/0956—Predicting travel path or likelihood of collision the prediction being responsive to traffic or environmental parameters
-
- 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
- B60W2050/0028—Mathematical models, e.g. for simulation
- B60W2050/0029—Mathematical model of the driver
-
- 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
- B60W2520/00—Input parameters relating to overall vehicle dynamics
- B60W2520/10—Longitudinal speed
-
- 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
- B60W2520/00—Input parameters relating to overall vehicle dynamics
- B60W2520/12—Lateral speed
-
- 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
- B60W2554/00—Input parameters relating to objects
-
- 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
- B60W2710/00—Output or target parameters relating to a particular sub-units
- B60W2710/18—Braking system
-
- 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
- B60W2710/00—Output or target parameters relating to a particular sub-units
- B60W2710/20—Steering systems
Abstract
It is including as follows the invention discloses a kind of unmanned vehicle lane-change paths planning method for imitating outstanding driver：1：The outstanding pilot model based on genetic optimization BP neural network is established, is trained using a large amount of lane-change experimental datas, model is reached predetermined accuracy；2：Environmental information is obtained according to environmental perception module, as track is wide, Obstacle Position, Current vehicle travel speed is obtained according to vehicle speed sensor；3：According to the Obstacle Position and speed information obtained in 2, lane-changing intention is determined；4：Using Obstacle Position, speed, lane-changing intention and driving style as input, the Part I of lane-change driving path under output current working is calculated by outstanding pilot model；5：According to the Part I lane-change driving path obtained in step 4, Part I path is subjected to 180 ° of rotation around tie point, obtains whole lane-change path.The present invention cooks up the lane-change path of imitative outstanding driver's traveling, improves stability and comfortableness according to different operating modes.
Description
Technical Field
The invention relates to the technical field of unmanned vehicles, in particular to a road changing path planning method for an unmanned vehicle.
Background
With the development of society, the vehicle holding amount is increased sharply, and along with the increase, frequent accidents cause property loss. Accidents are mainly caused by improper operation of the driver. The unmanned vehicle can remove the unstable factor of 'people', and the vehicle is directly controlled by a computer, so that the driving safety is greatly improved. As a result, unmanned technology has gained increasing attention.
The path planning is one of key technologies of unmanned driving, and can plan a path which can enable the unmanned vehicle to safely and stably run according to information such as the running speed of the vehicle, the environmental conditions and the like.
According to research, when the intelligent automobile is taken, the proportion of people with car sickness is greatly higher than that of people taking a traditional automobile. The reason for this is that the route traveled by the unmanned vehicle is greatly different from the route traveled by the actual driver, and the vehicle cannot be smoothly driven during steering. Therefore, there is a need for an improved method for planning the path of an unmanned vehicle, which improves the driving stability of the vehicle.
National patent 201110007154.X proposes to use a gravitational point function of an artificial potential field method to plan a driving path of an unmanned vehicle, but the curvature of the path at a turning position is discontinuous, so that the driving stability of the vehicle is influenced.
When driving a vehicle, an excellent driver plans an optimal driving path according to vehicle information and environment information, and then operates a steering wheel, an accelerator pedal or a brake pedal to enable the vehicle to stably drive along the planned path. Although the unmanned vehicle may plan a path based on the relevant information using some advanced algorithms. However, the path generated by the advanced algorithm is very different from the path actually traveled by the driver, which makes the unmanned vehicle less comfortable. By learning the driving path of the excellent driver under different working conditions, the unmanned vehicle can smoothly drive as if driven by the excellent driver.
Disclosure of Invention
In order to improve the comfort and stability of the unmanned vehicle, the invention provides an unmanned vehicle lane change path planning method simulating an excellent driver, which can plan a lane change driving path simulating the excellent driver to enable the unmanned vehicle to run more stably according to vehicle running information and environment information.
The technical scheme adopted by the invention for solving the technical problem is as follows:
an unmanned vehicle road changing path planning method imitating an excellent driver comprises the following steps:
step 1: an excellent driver model based on a genetic optimization BP neural network is established, and a large amount of lane change experimental data is used for training to enable the model to reach preset precision.
Step 2: and obtaining environmental information such as lane width and barrier positions according to the environment sensing module, and obtaining the current vehicle running speed according to the vehicle speed sensor.
And 3, step 3: and (3) determining the lane changing intention according to the obstacle position and the vehicle speed information obtained in the step (2).
And 4, step 4: and outputting a first part of the lane changing driving path under the current working condition through calculation of an excellent driver model by taking the position of the obstacle, the speed of the vehicle, the lane changing intention and the driving type as input.
And 5: and 4, rotating the first part of the path around the connecting point by 180 degrees according to the first part of the driving path obtained in the step 4 to obtain the whole road changing path.
Further, the step 1 is specifically as follows:
step 1.1, inviting five experienced drivers to carry out lane change real-time tests, and dividing the five drivers into an aggressive type, a middle type and a conservative type according to the driving characteristics of the drivers. The test vehicle is equipped with a GPS device for recording the actual driving path. And respectively requesting each driver to maintain the speed of 20km/h, 30km/h, 40km/h and 50km/h, carrying out lane changing tests with different lane changing intentions and different positions of obstacles, and recording a driving path. Each group of tests is carried out for three times, and an optimal group of data is selected as final test data under the working condition.
Step 1.2, dividing the lane-changing driving path into two parts according to the input angle of the steering wheel, wherein the path from the beginning of steering to the time when the steering wheel angle is zero for the first time is the first part, and the rest driving path is the second part.
Step 1.3, fitting the first part of the path by using a cubic polynomial to ensure that the starting point of the path is superposed with the origin point of a coordinate system to obtain a path function expression of the first part:
wherein a is 1 *、a 2 *、a 3 * Is a polynomial coefficient and is obtained by function fitting.
Step 1.4, a genetic optimization BP neural network model is established, the driving type, the vehicle speed, the lane changing intention and the position of an obstacle are input, various coefficients of a cubic function expression of a first part path are output, training is carried out by using test data, and an excellent driver model based on the genetic optimization BP neural network is established.
Further, the step 3 is specifically as follows:
and 3.1, obtaining the position of the obstacle according to the environment sensing module, calculating the distance D between the obstacle and the vehicle, and obtaining the vehicle speed v at the moment according to the vehicle speed sensor.
Step 3.2, if D is more than 60m or 30m is formed by woven fabric D and woven fabric 60m, and v is less than 40km/h, the common track changing purpose is achieved; the rest are the urgent obstacle avoidance and track change intentions.
Further, the step 4 is specifically as follows:
and 4.1, determining the driving type according to the selection of the passenger.
Step 4.2, the obstacle position and the vehicle speed obtained according to the step 2, the lane changing intention obtained according to the step 3, the driving type determined in the step 4.1 and other four parameters are used as input, an excellent driver model based on the genetic optimization BP neural network obtained in the step 1 is calculated, and a first part path expression simulating lane changing of an excellent driver is obtained as follows:
y＝a 3 x 3 +a 2 x 2 +a 1 x (2)
wherein a is 1 、a 2 、a 3 The coefficient is a path cubic polynomial coefficient and is obtained by genetic optimization BP neural network.
Further, the step 5 is specifically as follows:
and 5.1, obtaining the lane width of the driving road to be 2B according to the environment sensing module.
And 5.2, according to the first path switching expression obtained in the step 4, rotating the first path around the connecting point by 180 degrees to obtain a second path in order to enable the tangential directions of the two paths at the connecting point to be the same.
Step 5.3, according to the lane width obtained in step 5.1 and the second partial path function expression obtained in step 5.2, the whole road changing path function expression is as follows:
wherein a is satisfying the equation B = a 3 ·a 3 +a 2 ·a 2 +a 1 The value of a.
The invention has the beneficial effects that:
according to the invention, through a large number of real vehicle lane changing tests, influence factors of actual lane changing paths are researched, and an excellent driver model is established. Can plan out the route that trades that an imitative outstanding driver went according to the operating mode of traveling of difference, can improve the stability of traveling of vehicle, increase passenger's travelling comfort.
Drawings
Fig. 1 is a configuration diagram of a road change path planning method for a simulated excellent driver.
FIG. 2 is a flow chart of a genetic optimization BP neural network model.
Fig. 3 is a schematic diagram of a lane-change travel path.
Detailed Description
The invention is further described with reference to the following figures and specific examples.
As shown in fig. 1, the present invention provides a method for planning a road diameter of an unmanned vehicle imitating an excellent driver, comprising the following steps:
step 1: an excellent driver model based on a genetic optimization BP neural network is established, and a large amount of lane change experimental data are used for training to enable the model to reach preset precision.
Further, the step 1 is specifically as follows:
step 1.1, inviting five experienced drivers to carry out lane change real-time tests, and dividing the five drivers into an aggressive type, a middle type and a conservative type according to the driving characteristics of the drivers. The test vehicle is equipped with a GPS device for recording the actual travel path. And respectively requesting each driver to maintain the speed of 20km/h, 30km/h, 40km/h and 50km/h, carrying out lane change tests with different lane change intentions and different barrier positions, and recording the driving path. Each set of test is carried out three times, and an optimal set of data is selected as final test data under the working condition.
Step 1.2, dividing the lane-changing driving path into two parts according to the input angle of the steering wheel, wherein the path from the beginning of steering to the time when the steering wheel angle is zero for the first time is the first part, and the rest driving path is the second part.
And 1.3, establishing a vehicle coordinate system, wherein the center of mass of the vehicle when the vehicle starts to turn is taken as an origin, the advancing direction of the vehicle is the positive direction of an x axis, and the positive direction of a y axis points to the left side of a driver. Fitting the first part of the path by using a cubic polynomial to enable the starting point of the path to coincide with the origin of a coordinate system to obtain a first part path function expression:
step 1.4, a genetic optimization BP neural network model is established, the driving type, the vehicle speed, the lane changing intention and the position of an obstacle are input, various coefficients of a first part path cubic function expression are output, training is carried out by using test data, and an excellent driver model based on the genetic optimization BP neural network is established.
Fig. 2 shows a flowchart of a genetic optimization BP neural network model, which specifically includes the following steps:
step 1.4.1 determining the topological structure of the neural network: the number m of hidden layer nodes, the number n of input layer nodes, and the number l of output layer nodes.
Step 1.4.2 initializing the genetic algorithm population, generating an initial population X = (X) of size P 1 ,X 2 ,…,X p ) T . Obtaining population individual X by adopting a real number coding mode according to the topological structure of the neural network i ＝(x 1 ,x 2 ,…,x s ) As a chromosome in a genetic algorithm. The length of the chromosome is
s＝n×m+m×l+m+l (4)
In the formula, m is the number of hidden layer nodes, n is the number of input layer nodes, and l is the number of output layer nodes.
Step 1.4.3, determining a fitness function, assigning values to the connection weight and the threshold value of the BP neural network by using a chromosome, inputting a sample to train the neural network, and obtaining a network output value o i Then individual X in the population i Has a fitness function of
In the formula, y j K is a coefficient for the desired output training value.
Step 1.4.4 selection operation, the invention adopts roulette method to select operators, each individual X i Probability of being selected p i Comprises the following steps:
f i ＝k/F i (6)
step 1.4.5, performing crossover operation, wherein the crossover operation of the kth chromosome and the l chromosome at the jth gene is as follows:
wherein x is kj Is the j gene of the k chromosome, x lj Is the j-th gene of the l-th chromosome, b is [0,1]Random number in between.
Step 1.4.6 mutation operation, wherein the mutation operation of j gene of i chromosome is as follows:
wherein x is min And x max Are respectively based on x ij R is [0,1]]Random number of cells r 2 Is a random number, G is the current iteration number, G max For maximum number of evolutions, x ij Is the j gene of the ith chromosome.
And 1.4.7, calculating a fitness value, judging whether a termination condition is met, and obtaining the optimal initial weight and threshold of the BP neural network. And adjusting the weight and the threshold value by judging whether the error value of the BP neural network meets the precision requirement, so as to obtain the optimal network weight and threshold value.
Step 2: and obtaining environmental information such as lane width and barrier positions according to the environment sensing module, and obtaining the current vehicle running speed according to the vehicle speed sensor.
And step 3: and (4) determining the lane changing intention according to the obstacle position and the vehicle speed information obtained in the step (2).
Further, the step 3 is specifically as follows:
and 3.1, obtaining the position of the obstacle according to the environment sensing module, calculating the distance D between the obstacle and the vehicle, and obtaining the vehicle speed v at the moment according to the vehicle speed sensor.
Step 3.2, if D is more than 60m or 30m is formed by woven fabric D and woven fabric 60m, and v is less than 40km/h, the common track changing purpose is achieved; the rest are the urgent obstacle avoidance and track change purposes.
And 4, step 4: and outputting a first part of the lane changing driving path under the current working condition through calculation of an excellent driver model by taking the position of the obstacle, the speed of the vehicle, the lane changing intention and the driving type as input.
Further, the step 4 is specifically as follows:
and 4.1, determining the driving type according to the selection of the passenger.
Step 4.2, taking the obstacle position and the vehicle speed obtained in the step 2, the lane change intention obtained in the step 3, the driving type determined in the step 4.1 and other four parameters as input, and calculating the excellent driver model based on the genetic optimization BP neural network obtained in the step 1 to obtain a first part path expression simulating the lane change of the excellent driver, wherein the first part path expression comprises the following steps:
y＝a 3 x 3 +a 2 x 2 +a 1 x
and 5: and 4, rotating the first part of the path around the connecting point by 180 degrees according to the first part of the driving path obtained in the step 4 to obtain the whole road changing path.
Further, the step 5 is specifically as follows:
and 5.1, obtaining the lane width of the driving road to be 2B according to the environment sensing module.
And 5.2, according to the lane change first partial path expression obtained in the step 4, rotating the first partial path by 180 degrees around the connecting point to obtain a second partial path in order to enable the tangential directions of the two partial paths at the connecting point to be the same. Fig. 3 is a schematic diagram of the lane-changing travel path.
And 5.3, according to the lane width obtained in the step 5.1 and the second partial path function expression obtained in the step 5.2, the whole road changing path function expression is as follows:
wherein B = a 3 ·a 3 +a 2 ·a 2 +a 1 A. Wherein a is satisfying the equation B = a 3 ·a 3 +a 2 ·a 2 +a 1 The value of a.
The above-listed detailed description is only a specific description of a possible embodiment of the present invention, and they are not intended to limit the scope of the present invention, and equivalent embodiments or modifications made without departing from the technical spirit of the present invention should be included in the scope of the present invention.
Claims (9)
1. An unmanned vehicle road changing path planning method simulating an excellent driver is characterized by comprising the following steps:
step 1: establishing an excellent driver model based on a genetic optimization BP neural network, and training by using a large amount of lane change experimental data to enable the model to reach a preset precision;
and 2, step: obtaining environmental information such as lane width and barrier position according to an environmental sensing module, and obtaining the current vehicle running speed according to a vehicle speed sensor;
and step 3: determining a lane changing intention according to the obstacle position and the vehicle speed information obtained in the step 2;
and 4, step 4: taking the position of an obstacle, the speed of a vehicle, the lane changing intention and the driving type as input, and calculating and outputting a first part of a lane changing driving path under the current working condition through an excellent driver model;
and 5: and 4, rotating the first part of the path by 180 degrees around the connecting point according to the first part of the lane changing driving path obtained in the step 4 to obtain the whole lane changing path.
2. The method for planning the unmanned vehicle change road path imitating excellent drivers as claimed in claim 1, wherein the specific process of the step 1 comprises the following steps:
step 1.1, inviting a plurality of experienced drivers to carry out lane change real vehicle tests, and simultaneously recording actual lane change driving paths under different working conditions;
step 1.2, dividing the actual lane-changing driving path into two parts according to the steering wheel input angle according to the actual lane-changing driving path recorded in the step 1.1, wherein the path from the beginning of steering to the moment when the steering wheel angle is zero for the first time is a first part, and the rest driving path is a second part;
step 1.3, fitting the first part of the path to enable the starting point of the path to coincide with the origin of a coordinate system, and obtaining a path function expression of the first part of the actual driving:
step 1.4, a genetic optimization BP neural network model is established, the driving type, the vehicle speed, the lane changing intention and the position of an obstacle are input, various coefficients of a first part path cubic function expression are output, training is carried out by using test data, and an excellent driver model based on the genetic optimization BP neural network is established.
3. The method as claimed in claim 2, wherein the step 1.1 of performing the lane change real vehicle test specifically comprises: according to the driving characteristics of drivers, a plurality of drivers are divided into an aggressive type, a medium type and a conservative type; the test vehicle is provided with a GPS device for recording an actual driving path; respectively requesting each driver to maintain the speed of 20km/h, 30km/h, 40km/h and 50km/h, carrying out lane change tests according to different lane change intentions and different barrier positions, and recording a driving path; each group of tests is carried out for a plurality of times, and an optimal group of data is selected as final test data under the working condition.
4. The method as claimed in claim 2, wherein a cubic polynomial is used for the first part of the route in step 1.3, and the first part of the route function expression is:
5. the method for planning the unmanned vehicle change road path imitating excellent drivers in claim 2, wherein the method for establishing the genetic optimization BP neural network model in the step 1.4 comprises the following steps: and (3) assigning the initial weight and the threshold value of the BP neural network by using the optimal individual obtained by the genetic algorithm, and carrying out local optimization by using a BP neural network prediction model so as to obtain a BP neural network prediction value with a global optimal solution.
6. The method as claimed in claim 5, wherein the genetic optimization BP neural network model in step 1.4 is implemented as follows:
step 1.4.1 determining the topological structure of the neural network: the number m of hidden layer nodes, the number n of input layer nodes and the number l of output layer nodes;
step 1.4.2 initializing the genetic algorithm population, generating an initial population X = (X) with the scale of P 1 ,X 2 ,…,X p ) T, obtaining population individuals X by adopting a real number coding mode according to the topological structure of the neural network i ＝(x 1 ,x 2 ,…,x s ) As a chromosome in the genetic algorithm, the length of the chromosome is
s＝n×m+m×l+m+l
Step 1.4.3 is to determine the fitness function, the chromosome is used for carrying out assignment on the connection weight and the threshold value of the BP neural network, the input sample is used for carrying out neural network training, and a network output value o is obtained j Then individual X in the population i Has a fitness function of
In the formula, y j K is a coefficient for the desired output training value.
Step 1.4.4 selection operation, selecting operators by roulette, for each individual X i Probability of being selected p i Comprises the following steps:
f i ＝k/F i
step 1.4.5, performing crossover operation, wherein the crossover operation of the kth chromosome and the l chromosome at the jth gene is as follows:
wherein b is a random number between [0,1 ].
Step 1.4.6 mutation operation, wherein the mutation operation of j gene of i chromosome is as follows:
wherein x is min And x max Are respectively based on x ij R is [0,1]]Random number of cells r 2 Is oneRandom number, G is the current iteration number, G max Is the maximum number of evolutions;
and 1.4.7, calculating a fitness value, judging whether a termination condition is met, and obtaining the optimal initial weight and threshold of the BP neural network. And adjusting the weight and the threshold value by judging whether the error value of the BP neural network meets the precision requirement or not to obtain the optimal network weight and the optimal threshold value.
7. The method for planning the unmanned vehicle change road path imitating excellent drivers as claimed in claim 1, wherein the specific process of the step 3 comprises the following steps:
3.1, obtaining the position of the obstacle according to the environment sensing module, calculating the distance D between the obstacle and the vehicle, and obtaining the vehicle speed v according to the vehicle speed sensor;
step 3.2, if D is more than 60m or 30m is formed by woven fabric D and woven fabric 60m, and v is less than 40km/h, the common track changing purpose is achieved; if D <60m and v >40km/h or D <30m, the lane change intention is urgent obstacle avoidance.
8. The method for planning the unmanned vehicle change road path imitating excellent drivers as claimed in claim 1, wherein the specific process of the step 4 comprises the following steps:
step 4.1, determining the driving type according to the selection of the passenger;
step 4.2, taking the obstacle position and the vehicle speed obtained in the step 2, the lane change intention obtained in the step three, the driving type determined in the step 4.1 and other four parameters as input, and calculating the excellent driver model based on the genetic optimization BP neural network obtained in the step 1 to obtain a first part path expression simulating the lane change of the excellent driver, wherein the first part path expression comprises the following steps:
y＝a 3 x 3 +a 2 x 2 +a 1 x。
9. the method for planning the unmanned vehicle change road path imitating excellent drivers as claimed in claim 1, wherein the specific process of the step 5 comprises the following steps:
step 5.1, obtaining the lane width of the driving road to be 2B according to the environment sensing module;
step 5.2, according to the lane change first part path expression obtained in the step 4, in order to enable the tangent directions of the two part paths at the connecting point to be the same, rotating the first part path by 180 degrees around the connecting point to obtain a second part path;
and 5.3, according to the lane width obtained in the step 5.1 and the second partial path function expression obtained in the step 5.2, the whole road changing path function expression is as follows:
wherein B = a 3 ·a 3 +a 2 ·a 2 +a 1 A. Wherein a is a value satisfying the equation B = a 3 ·a 3 +a 2 ·a 2 +a 1 The value of a.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710953896.9A CN107813820A (en) | 2017-10-13 | 2017-10-13 | A kind of unmanned vehicle lane-change paths planning method for imitating outstanding driver |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710953896.9A CN107813820A (en) | 2017-10-13 | 2017-10-13 | A kind of unmanned vehicle lane-change paths planning method for imitating outstanding driver |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107813820A true CN107813820A (en) | 2018-03-20 |
Family
ID=61607953
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710953896.9A Pending CN107813820A (en) | 2017-10-13 | 2017-10-13 | A kind of unmanned vehicle lane-change paths planning method for imitating outstanding driver |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107813820A (en) |
Cited By (19)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109283843A (en) * | 2018-10-12 | 2019-01-29 | 江苏大学 | A kind of lane-change method for planning track merged based on multinomial with particle swarm algorithm |
| CN109367541A (en) * | 2018-10-15 | 2019-02-22 | 吉林大学 | A kind of intelligent vehicle class people's lane change decision-making technique based on driver behavior pattern |
| CN109739218A (en) * | 2018-12-24 | 2019-05-10 | 江苏大学 | It is a kind of that outstanding driver's lane-change method for establishing model is imitated based on GRU network |
| WO2019206513A1 (en) * | 2018-04-27 | 2019-10-31 | Bayerische Motoren Werke Aktiengesellschaft | Method for driving manouevre assistance of a vehicle, device, computer program, and computer program product |
| CN110525446A (en) * | 2019-09-06 | 2019-12-03 | 山东理工大学 | A kind of automobile pressure lane-change decision safe early warning method considering mood |
| CN110956851A (en) * | 2019-12-02 | 2020-04-03 | 清华大学 | Intelligent networking automobile cooperative scheduling lane changing method |
| CN111256714A (en) * | 2018-11-30 | 2020-06-09 | 北汽福田汽车股份有限公司 | Path planning method and device and vehicle |
| WO2020132954A1 (en) * | 2018-12-26 | 2020-07-02 | Baidu.Com Times Technology (Beijing) Co., Ltd. | Optimal planner switch method for three point turn of autonomous driving vehicles |
| CN111581887A (en) * | 2020-05-16 | 2020-08-25 | 郑州轻工业大学 | Unmanned vehicle intelligent training method based on simulation learning in virtual environment |
| CN111796587A (en) * | 2019-03-21 | 2020-10-20 | 北京京东尚科信息技术有限公司 | Automatic driving method, storage medium and electronic device |
| CN111845787A (en) * | 2020-08-03 | 2020-10-30 | 北京理工大学 | Lane change intention prediction method based on LSTM |
| CN112099515A (en) * | 2020-11-16 | 2020-12-18 | 北京鼎翰科技有限公司 | Automatic driving method for lane change avoidance |
| CN112149796A (en) * | 2020-08-13 | 2020-12-29 | 江苏大学 | Driving style identification method for optimizing BP neural network based on improved genetic algorithm |
| CN113184040A (en) * | 2021-06-03 | 2021-07-30 | 长安大学 | Unmanned vehicle line-controlled steering control method and system based on steering intention of driver |
| CN114115234A (en) * | 2021-10-28 | 2022-03-01 | 江苏大学 | Unmanned vehicle road change path planning method based on monitoring strategy |
| US11279350B2 (en) | 2019-12-27 | 2022-03-22 | Automotive Research & Testing Center | Method of adaptive trajectory generation for a vehicle |
| CN114364592A (en) * | 2019-06-10 | 2022-04-15 | Lit汽车公司 | Method and apparatus for trajectory shape generation for autonomous vehicles |
| CN114613142A (en) * | 2022-03-24 | 2022-06-10 | 长沙理工大学 | Rule-based automatic driving intersection vehicle lane change control method |
| CN115457783A (en) * | 2022-08-30 | 2022-12-09 | 重庆长安汽车股份有限公司 | Method and system for traffic, cooperation and cooperation at signal lamp-free intersection |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106740457A (en) * | 2016-12-07 | 2017-05-31 | 镇江市高等专科学校 | Vehicle lane-changing decision-making technique based on BP neural network model |
| CN106874597A (en) * | 2017-02-16 | 2017-06-20 | 北理慧动（常熟）车辆科技有限公司 | A kind of highway passing behavior decision-making technique for being applied to automatic driving vehicle |
| CN106873584A (en) * | 2017-01-11 | 2017-06-20 | 江苏大学 | Pilotless automobile apery turns to the method for building up of rule base |
| CN107161155A (en) * | 2017-04-27 | 2017-09-15 | 大连理工大学 | A kind of vehicle collaboration lane-change method and its system based on artificial neural network |
-
2017
- 2017-10-13 CN CN201710953896.9A patent/CN107813820A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106740457A (en) * | 2016-12-07 | 2017-05-31 | 镇江市高等专科学校 | Vehicle lane-changing decision-making technique based on BP neural network model |
| CN106873584A (en) * | 2017-01-11 | 2017-06-20 | 江苏大学 | Pilotless automobile apery turns to the method for building up of rule base |
| CN106874597A (en) * | 2017-02-16 | 2017-06-20 | 北理慧动（常熟）车辆科技有限公司 | A kind of highway passing behavior decision-making technique for being applied to automatic driving vehicle |
| CN107161155A (en) * | 2017-04-27 | 2017-09-15 | 大连理工大学 | A kind of vehicle collaboration lane-change method and its system based on artificial neural network |
Non-Patent Citations (1)
| Title |
|---|
| 安林芳: "智能车辆自动驾驶路径规划研究", 《CNKI优秀硕士学位论文全文库》 * |
Cited By (24)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2019206513A1 (en) * | 2018-04-27 | 2019-10-31 | Bayerische Motoren Werke Aktiengesellschaft | Method for driving manouevre assistance of a vehicle, device, computer program, and computer program product |
| US11820379B2 (en) | 2018-04-27 | 2023-11-21 | Bayerische Motoren Werke Aktiengesellschaft | Method for driving maneuver assistance of a vehicle, device, computer program, and computer program product |
| CN109283843A (en) * | 2018-10-12 | 2019-01-29 | 江苏大学 | A kind of lane-change method for planning track merged based on multinomial with particle swarm algorithm |
| 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 |
| CN109367541A (en) * | 2018-10-15 | 2019-02-22 | 吉林大学 | A kind of intelligent vehicle class people's lane change decision-making technique based on driver behavior pattern |
| CN111256714A (en) * | 2018-11-30 | 2020-06-09 | 北汽福田汽车股份有限公司 | Path planning method and device and vehicle |
| CN109739218A (en) * | 2018-12-24 | 2019-05-10 | 江苏大学 | It is a kind of that outstanding driver's lane-change method for establishing model is imitated based on GRU network |
| WO2020132954A1 (en) * | 2018-12-26 | 2020-07-02 | Baidu.Com Times Technology (Beijing) Co., Ltd. | Optimal planner switch method for three point turn of autonomous driving vehicles |
| CN111796587A (en) * | 2019-03-21 | 2020-10-20 | 北京京东尚科信息技术有限公司 | Automatic driving method, storage medium and electronic device |
| CN114364592A (en) * | 2019-06-10 | 2022-04-15 | Lit汽车公司 | Method and apparatus for trajectory shape generation for autonomous vehicles |
| CN110525446A (en) * | 2019-09-06 | 2019-12-03 | 山东理工大学 | A kind of automobile pressure lane-change decision safe early warning method considering mood |
| CN110956851B (en) * | 2019-12-02 | 2020-11-24 | 清华大学 | Intelligent networking automobile cooperative scheduling lane changing method |
| CN110956851A (en) * | 2019-12-02 | 2020-04-03 | 清华大学 | Intelligent networking automobile cooperative scheduling lane changing method |
| US11279350B2 (en) | 2019-12-27 | 2022-03-22 | Automotive Research & Testing Center | Method of adaptive trajectory generation for a vehicle |
| CN111581887B (en) * | 2020-05-16 | 2023-04-07 | 郑州轻工业大学 | Unmanned vehicle intelligent training method based on simulation learning in virtual environment |
| CN111581887A (en) * | 2020-05-16 | 2020-08-25 | 郑州轻工业大学 | Unmanned vehicle intelligent training method based on simulation learning in virtual environment |
| CN111845787A (en) * | 2020-08-03 | 2020-10-30 | 北京理工大学 | Lane change intention prediction method based on LSTM |
| CN112149796A (en) * | 2020-08-13 | 2020-12-29 | 江苏大学 | Driving style identification method for optimizing BP neural network based on improved genetic algorithm |
| CN112099515A (en) * | 2020-11-16 | 2020-12-18 | 北京鼎翰科技有限公司 | Automatic driving method for lane change avoidance |
| CN113184040A (en) * | 2021-06-03 | 2021-07-30 | 长安大学 | Unmanned vehicle line-controlled steering control method and system based on steering intention of driver |
| CN114115234A (en) * | 2021-10-28 | 2022-03-01 | 江苏大学 | Unmanned vehicle road change path planning method based on monitoring strategy |
| CN114613142A (en) * | 2022-03-24 | 2022-06-10 | 长沙理工大学 | Rule-based automatic driving intersection vehicle lane change control method |
| CN115457783A (en) * | 2022-08-30 | 2022-12-09 | 重庆长安汽车股份有限公司 | Method and system for traffic, cooperation and cooperation at signal lamp-free intersection |
| CN115457783B (en) * | 2022-08-30 | 2023-08-11 | 重庆长安汽车股份有限公司 | Traffic, cooperation and cooperation method and system for intersection without signal lamp |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107813820A (en) | A kind of unmanned vehicle lane-change paths planning method for imitating outstanding driver | |
| Huang et al. | Personalized trajectory planning and control of lane-change maneuvers for autonomous driving | |
| CN107340772B (en) | Unmanned-oriented anthropomorphic reference trajectory planning method | |
| CN108256233B (en) | Intelligent vehicle trajectory planning and tracking method and system based on driver style | |
| Zhang et al. | A game theoretic model predictive controller with aggressiveness estimation for mandatory lane change | |
| CN106740846B (en) | A kind of electric car self-adapting cruise control method of double mode switching | |
| Pérez et al. | Trajectory generator for autonomous vehicles in urban environments | |
| CN110304074B (en) | Hybrid driving method based on layered state machine | |
| CN111681452B (en) | Unmanned vehicle dynamic lane change track planning method based on Frenet coordinate system | |
| Lattarulo et al. | Urban motion planning framework based on n-bézier curves considering comfort and safety | |
| CN110928297A (en) | Intelligent bus route planning method based on multi-objective dynamic particle swarm optimization | |
| CN114312830A (en) | Intelligent vehicle coupling decision model and method considering dangerous driving conditions | |
| Jayawardana et al. | Learning eco-driving strategies at signalized intersections | |
| Hu et al. | Adaptive lane change trajectory planning scheme for autonomous vehicles under various road frictions and vehicle speeds | |
| Torabi et al. | Energy minimization for an electric bus using a genetic algorithm | |
| Li et al. | Human-like trajectory planning on curved road: Learning from human drivers | |
| CN110956851A (en) | Intelligent networking automobile cooperative scheduling lane changing method | |
| CN113715842A (en) | High-speed moving vehicle control method based on simulation learning and reinforcement learning | |
| Gim et al. | Safe and efficient lane change maneuver for obstacle avoidance inspired from human driving pattern | |
| Sun et al. | Human-like highway trajectory modeling based on inverse reinforcement learning | |
| CN116118780A (en) | Vehicle obstacle avoidance track planning method, system, vehicle and storage medium | |
| CN115257819A (en) | Decision-making method for safe driving of large-scale commercial vehicle in urban low-speed environment | |
| CN115257789A (en) | Decision-making method for side anti-collision driving of commercial vehicle in urban low-speed environment | |
| Liu et al. | Mtd-gpt: A multi-task decision-making gpt model for autonomous driving at unsignalized intersections | |
| CN114030485A (en) | Automatic driving automobile man lane change decision planning method considering attachment coefficient |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20180320 |