CN112835362B - Automatic lane change planning method and device, electronic equipment and storage medium - Google Patents
Automatic lane change planning method and device, electronic equipment and storage medium Download PDFInfo
- Publication number
- CN112835362B CN112835362B CN202011593899.4A CN202011593899A CN112835362B CN 112835362 B CN112835362 B CN 112835362B CN 202011593899 A CN202011593899 A CN 202011593899A CN 112835362 B CN112835362 B CN 112835362B
- Authority
- CN
- China
- Prior art keywords
- lane change
- simulation model
- vehicle simulation
- path
- model
- 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
Links
- 230000008859 change Effects 0.000 title claims abstract description 270
- 238000000034 method Methods 0.000 title claims abstract description 49
- 238000004088 simulation Methods 0.000 claims abstract description 148
- 230000007613 environmental effect Effects 0.000 claims abstract description 31
- 238000012549 training Methods 0.000 claims description 19
- 230000009471 action Effects 0.000 claims description 18
- 238000004590 computer program Methods 0.000 claims description 9
- 230000015654 memory Effects 0.000 claims description 9
- 230000008447 perception Effects 0.000 claims description 7
- 238000013433 optimization analysis Methods 0.000 claims description 5
- 238000003062 neural network model Methods 0.000 abstract description 7
- 238000004891 communication Methods 0.000 description 5
- 230000008569 process Effects 0.000 description 4
- 238000013527 convolutional neural network Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- 230000008901 benefit Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 230000003044 adaptive effect Effects 0.000 description 1
- 230000006399 behavior Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000005012 migration Effects 0.000 description 1
- 238000013508 migration Methods 0.000 description 1
- 238000003825 pressing Methods 0.000 description 1
- 230000000306 recurrent effect Effects 0.000 description 1
- 230000002787 reinforcement Effects 0.000 description 1
- 230000006403 short-term memory Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
- G05D1/02—Control of position or course in two dimensions
- G05D1/021—Control of position or course in two dimensions specially adapted to land vehicles
- G05D1/0231—Control of position or course in two dimensions specially adapted to land vehicles using optical position detecting means
- G05D1/0242—Control of position or course in two dimensions specially adapted to land vehicles using optical position detecting means using non-visible light signals, e.g. IR or UV signals
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
- G05D1/02—Control of position or course in two dimensions
- G05D1/021—Control of position or course in two dimensions specially adapted to land vehicles
- G05D1/0212—Control of position or course in two dimensions specially adapted to land vehicles with means for defining a desired trajectory
- G05D1/0214—Control of position or course in two dimensions specially adapted to land vehicles with means for defining a desired trajectory in accordance with safety or protection criteria, e.g. avoiding hazardous areas
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
- G05D1/02—Control of position or course in two dimensions
- G05D1/021—Control of position or course in two dimensions specially adapted to land vehicles
- G05D1/0212—Control of position or course in two dimensions specially adapted to land vehicles with means for defining a desired trajectory
- G05D1/0221—Control of position or course in two dimensions specially adapted to land vehicles with means for defining a desired trajectory involving a learning process
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
- G05D1/02—Control of position or course in two dimensions
- G05D1/021—Control of position or course in two dimensions specially adapted to land vehicles
- G05D1/0212—Control of position or course in two dimensions specially adapted to land vehicles with means for defining a desired trajectory
- G05D1/0223—Control of position or course in two dimensions specially adapted to land vehicles with means for defining a desired trajectory involving speed control of the vehicle
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
- G05D1/02—Control of position or course in two dimensions
- G05D1/021—Control of position or course in two dimensions specially adapted to land vehicles
- G05D1/0276—Control of position or course in two dimensions specially adapted to land vehicles using signals provided by a source external to the vehicle
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
- G06F30/27—Design optimisation, verification or simulation using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/084—Backpropagation, e.g. using gradient descent
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T10/00—Road transport of goods or passengers
- Y02T10/10—Internal combustion engine [ICE] based vehicles
- Y02T10/40—Engine management systems
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Automation & Control Theory (AREA)
- Remote Sensing (AREA)
- Radar, Positioning & Navigation (AREA)
- Aviation & Aerospace Engineering (AREA)
- Theoretical Computer Science (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- Software Systems (AREA)
- General Engineering & Computer Science (AREA)
- Biomedical Technology (AREA)
- Mathematical Physics (AREA)
- Health & Medical Sciences (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Life Sciences & Earth Sciences (AREA)
- Electromagnetism (AREA)
- Geometry (AREA)
- Computer Hardware Design (AREA)
- Medical Informatics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Traffic Control Systems (AREA)
- Steering Control In Accordance With Driving Conditions (AREA)
Abstract
The invention provides an automatic lane change planning method and device, electronic equipment and a storage medium, wherein the method comprises the following steps: controlling the vehicle simulation model to acquire environmental information in a current path of a simulation road; inputting the environmental information into an autonomous lane change model to obtain lane change decision data; adjusting the current path based on the lane change decision data to obtain a lane change path; generating control parameters based on the variable road diameter to control the vehicle simulation model to run along the variable road diameter, so that the variable road decision data is determined by adopting a neural network model, and the rough variable road direction is determined; and then determining a specific lane change adjustment strategy by adopting a quadratic programming mode to obtain a lane change path, thereby obtaining an accurate and reasonable lane change path.
Description
Technical Field
The present invention relates to the field of automatic driving technologies, and in particular, to an automatic lane change planning method and apparatus, an electronic device, and a storage medium.
Background
Autopilot refers to a driving style that relies on AI technology to automatically drive a vehicle without human involvement.
In the field of autopilot, automatic planning of a driving route is a relatively important technique. The automatic driving vehicle is influenced by the speed of the front vehicle in the running process to generate a phenomenon of 'pressing speed', so that the automatic driving vehicle cannot run at a set speed, and the economic benefit of vehicle running is reduced. In order to improve the running efficiency of the vehicle, the automatic driving vehicle should have autonomous lane changing capability, so that the automatic driving vehicle can maintain a set speed through reasonable lane changing under the condition of traffic flow.
In the prior art, the reasons for the lane change running are mainly concentrated on the obstacle avoidance layer, and the influence of the dynamic environment is ignored. However, in the actual driving environment, lane change is a relatively complex driving behavior, and many influencing factors are involved, such as that the front is in red light, or that the speed of the front vehicle is faster than that of the vehicle on the fly, so that lane change is not suitable in this case.
Therefore, how to make accurate and reasonable lane change path according to the environment information is a technical problem to be solved at present.
Disclosure of Invention
The invention provides an automatic lane change planning method and device, electronic equipment and a storage medium, which are used for solving the technical defects in the prior art.
The invention also provides an automatic lane change planning method which is applied to the vehicle simulation model, and the method comprises the following steps:
controlling the vehicle simulation model to acquire environmental information in a current path of a simulation road;
inputting the environmental information into an autonomous lane change model to obtain lane change decision data;
adjusting the current path based on the lane change decision data to obtain a lane change path;
and generating control parameters based on the variable road diameter so as to control the vehicle simulation model to run along the variable road diameter.
According to the automatic lane change planning method provided by the invention, the vehicle simulation model is controlled to acquire the environmental information in the current path of the simulation road, and the method comprises the following steps:
acquiring the state information of the surrounding vehicles and the state information of the surrounding vehicles in the current path of the simulation road through a perception module of the vehicle simulation model;
wherein, the week car state information includes: obstacle information, relative position, relative velocity, and relative orientation;
the self state information includes: engine speed, engine power, gear, vehicle speed, and steering wheel angle.
According to the automatic lane change planning method provided by the invention, the autonomous lane change model is trained by the following modes:
and taking the environment sample information as a training sample, taking a lane change decision label as a training label, inputting the training label into the autonomous lane change model, and carrying out back propagation training, wherein the lane change decision label comprises a left lane change label, a non-lane change label and a right lane change label.
According to the automatic lane change planning method provided by the invention, the environmental information is input into an autonomous lane change model to obtain lane change decision data, and the method comprises the following steps:
inputting the environment information into an autonomous lane change model to obtain a plurality of lane change decision data and probability values thereof, wherein the lane change decision data comprises left lane change, unchanged lane change and right lane change;
adjusting the current path based on the lane change decision data to obtain a lane change path, comprising:
according to pre-stored path library information, adjusting the current path based on a plurality of lane change decision data;
and carrying out optimization analysis on the adjusted paths to determine that the optimal path is a lane change path.
According to the automatic lane-changing planning method provided by the invention, control parameters are generated based on a lane-changing path so as to control the vehicle simulation model to run along the lane-changing path, and the automatic lane-changing planning method comprises the following steps:
the variable road diameter is sent to a controller of a vehicle simulation model, so that the controller generates input control parameters of the vehicle simulation model according to the variable road diameter;
and controlling the action of a control part of the vehicle simulation model according to the input control parameters so as to control the vehicle simulation model to run along the variable road diameter.
According to the automatic lane change planning method provided by the invention, the input control parameters comprise: the opening angle of the accelerator pedal, the rotation angle of the brake pedal, the rotation direction and the rotation angle of the steering wheel;
controlling the action of a control component of the vehicle simulation model according to the input control parameters, comprising:
controlling the action of an accelerator pedal of the vehicle simulation model according to the opening angle of the accelerator pedal;
controlling the action of a brake pedal of the vehicle simulation model according to the rotation angle of the brake pedal; and/or
And controlling the steering wheel of the vehicle simulation model to rotate according to the rotation direction and the rotation angle of the steering wheel.
According to the automatic lane change planning method provided by the invention, after the action of the control component of the vehicle simulation model is controlled according to the input control parameter, the automatic lane change planning method further comprises the following steps:
acquiring current self-state information of the vehicle simulation model through a perception module of the vehicle simulation model;
and inputting the current self-state information of the vehicle simulation model into the autonomous lane change model so as to continuously train the autonomous lane change model.
The invention also provides an automatic lane change planning device which is applied to a vehicle simulation model, and the device comprises:
the environment information acquisition module is used for controlling the vehicle simulation model to acquire environment information in the current path of the simulation road;
the lane change decision generation module is used for inputting the environmental information into an autonomous lane change model to obtain lane change decision data;
the lane change path adjusting module is used for adjusting the current path based on lane change decision data to obtain a lane change path;
and the lane change driving module is used for generating control parameters based on the lane change path so as to control the vehicle simulation model to drive along the lane change path.
The invention also provides an electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, the processor implementing the steps of the automatic lane change planning method as described in any of the above when executing the program.
The present invention also provides a non-transitory computer readable storage medium having stored thereon a computer program which, when executed by a processor, implements the steps of the automatic lane change planning method as described in any of the above.
According to the automatic lane change planning method and device, the simulation road conforming to the actual scene is established, so that the vehicle simulation model runs in the simulation road to determine the environment information, the success rate of the vehicle simulation model moving from the simulation environment to the real environment is improved, then the environment information is input into the autonomous lane change model to obtain lane change decision data, the current path is adjusted based on the lane change decision data to obtain a lane change path, and therefore the lane change decision data is determined by adopting the neural network model firstly, and the rough lane change direction is determined; and then determining a specific lane change adjustment strategy by adopting a quadratic programming mode to obtain a lane change path, thereby obtaining an accurate and reasonable lane change path.
Drawings
In order to more clearly illustrate the invention or the technical solutions of the prior art, the following description will briefly explain the drawings used in the embodiments or the description of the prior art, and it is obvious that the drawings in the following description are some embodiments of the invention, and other drawings can be obtained according to the drawings without inventive effort for a person skilled in the art.
FIG. 1 is a schematic flow chart of an automatic lane change planning method according to the present invention;
FIG. 2 is a second flow chart of the automatic lane change planning method according to the present invention;
FIG. 3 is a schematic diagram of an automatic lane change planning apparatus according to the present invention;
fig. 4 is a schematic structural diagram of an electronic device provided by the present invention.
Detailed Description
For the purpose of making the objects, technical solutions and advantages of the present invention more apparent, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings, and it is apparent that the described embodiments are some embodiments of the present invention, not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
First, a vehicle simulation model according to the present embodiment will be described.
The method of the embodiment is applicable to a vehicle simulation model, not a real vehicle, for example, a vehicle simulation model generated by using a truckim vehicle simulation simulator, so as to improve the success rate of the autonomous lane change model from the simulation environment to the real environment. The control elements, such as an accelerator pedal, a brake pedal, a steering wheel and the like, can be physical entity elements, are integrated in one platform, support a user to control the operation of the steering wheel, the clutch pedal, the accelerator pedal, the brake pedal, the gear and the like of the virtual vehicle to be tested, and output relevant control data to a simulation running module of the vehicle simulation model.
In addition, the vehicle simulation model is provided with a sensing module which can sense the state of the surrounding vehicle, such as relative position, relative speed and the like; in addition, the control data of the engine, such as the engine speed, the engine power, the gear, etc., can be obtained through the feedback unit.
The sensing module may be a variety of sensors, such as an infrared sensor, a laser sensor, and the like.
The embodiment of the invention discloses an automatic lane change planning method, which is applied to the following steps of:
101. and controlling the vehicle simulation model to acquire the environmental information in the current path of the simulation road.
In this embodiment, in order to improve the algorithm migration success rate, the present invention models a road section with gradient and curvature change in a real scene, and constructs a real simulation road. A simulated driving environment is simulated through the simulated road, and the vehicle simulation model is controlled to drive on the simulated road.
Specifically, in the driving process of the simulation road of the control vehicle simulation model, the state information of the surrounding vehicle and the state information of the surrounding vehicle in the current path of the simulation road are obtained through a perception module of the simulation road of the control vehicle.
Wherein, week car state information includes: obstacle information, relative position, relative speed, and relative orientation, and own state information includes: engine speed, engine power, gear, vehicle speed, and steering wheel angle.
102. And inputting the environmental information into the autonomous lane change model to obtain lane change decision data.
In this embodiment, a reinforcement learning algorithm is introduced by applying an autonomous lane change model to obtain lane change decision data according to environmental information.
The autonomous lane-changing model may be a convolutional neural network model, such as a CNN (Convolutional Neural Networks) model, or a recurrent neural network model, such as a Transformer model, an LSTM (Long Short-Term Memory) model, or the like.
Wherein the autonomous lane-changing model is trained by: and taking the environment sample information as a training sample, taking a lane change decision label as a training label, inputting the training label into the autonomous lane change model, and carrying out back propagation training, wherein the lane change decision label comprises a left lane change label, a non-lane change label and a right lane change label.
Taking a transducer model as an example, the transducer comprises an encoder and a decoder, environmental sample information is input into the encoder to be encoded to obtain an encoded vector, and then the encoded vector is input into the decoder to be decoded to obtain output lane change decision sample data; and comparing the output lane change decision sample data with the lane change decision label to calculate a loss value, and performing back propagation training to ensure that the loss value is smaller than a threshold value.
The finally obtained autonomous lane change model has the capability of outputting lane change decision data according to the environmental information. The lane change decision comprehensively considers relevant factors such as speed, oil consumption, safety, comfort and the like so as to ensure the rationality and economy of the lane change decision.
In a common driving scene, the lane change decision mainly includes: lane change to the left, lane change to the right, no lane change.
For example, in a more complex driving scenario, where the front lane is congested and the speed of the preceding vehicle is less than the speed of the vehicle simulation model, the vehicle simulation model is required to perform lane changes. For the left lane and the right lane, no vehicle can change lanes, but the right lane is a bus lane and is not driven in according to traffic rules.
The channel changing decision data is obtained through an autonomous channel changing model as follows: lane change to the left with a probability of 0.78; the probability is 0.22 when the vehicle changes lanes to the right, so as to realize the control of the lane changing to the left of the vehicle simulation model.
103. And adjusting the current path based on the lane change decision data to obtain a lane change path.
In this embodiment, after the lane change decision data is obtained by the decision layer located at the upper layer, the lane change decision data is only a lane change strategy of one strategy, and the specific lane change path is also to adjust the actual path based on the lane change decision data.
Therefore, in this step, a reasonable path needs to be further planned by the track planning layer at the bottom layer in a quadratic programming manner to realize the lane change decision at the upper layer.
In the process, the vehicle simulation model can completely follow the lane change decision data to carry out lane change planning, and can also carry out adaptive modification on the lane change decision data according to actual situations.
For example, lane change decision data indicates that the vehicle simulation model may change lanes to the left, but a lane on the left may cause inconvenience in lane change at the next intersection, and may be not far from the next intersection, and the lane change path may be obtained in a manner of not changing lanes and then changing lanes to the left at the next intersection.
104. And generating control parameters based on the variable road diameter so as to control the vehicle simulation model to run along the variable road diameter.
Specifically, step 104 includes: transmitting the variable road diameter to a controller of the vehicle simulation model so that the controller generates input control parameters of the vehicle simulation model according to the variable road diameter; and controlling the action of a control part of the vehicle simulation model according to the input control parameters so as to control the vehicle simulation model to run along the variable road diameter.
According to the automatic lane change planning method provided by the embodiment, the simulation road conforming to the actual scene is established, so that the vehicle simulation model runs in the simulation road to determine the environment information, the success rate of the vehicle simulation model moving from the simulation environment to the real environment is improved, then the environment information is input into the automatic lane change model to obtain lane change decision data, the current path is adjusted based on the lane change decision data to obtain a lane change path, and therefore the lane change decision data is determined by adopting the neural network model firstly, and the rough lane change direction is determined; and then determining a specific lane change adjustment strategy by adopting a quadratic programming mode to obtain a lane change path, thereby obtaining an accurate and reasonable lane change path.
The embodiment of the invention also discloses an automatic lane change planning method, which comprises the following steps of:
201. and controlling the vehicle simulation model to acquire the environmental information in the current path of the simulation road.
The simulated road is a road generated by modeling according to road sections with gradient and curvature changes in a real scene, so that a simulated running environment is provided for the vehicle simulation model.
The environment information includes information of the state of the vehicle, such as obstacle information, relative position, relative speed, relative direction, etc., of the vehicle simulation model, and also includes information of the state of the vehicle simulation model itself, such as engine speed, engine power, gear, vehicle speed, and steering wheel angle.
202. And inputting the environmental information into an autonomous lane change model to obtain a plurality of lane change decision data and probability values thereof.
The lane change decision data comprises a left lane change, a constant lane change and a right lane change.
The foregoing embodiments have been explained for the autonomous lane change model, and will not be described in detail herein.
According to the lane change decision data output by the autonomous lane change model, a rough lane change strategy of the vehicle simulation model can be determined. For example, the data output by the autonomous lane-change model includes lane-change to the left 0.75, lane-change to the right 0.21, and lane-change to the constant 0.04, which means that the probability that the vehicle simulation model should lane-change to the left is far greater than other lane-change decisions.
203. And according to the pre-stored path library information, adjusting the current path based on the multiple lane change decision data.
The path library stores a plurality of paths which can be acquired through a network in advance, so that alternative paths are provided for subsequent path adjustment. In addition, other information may be stored in the path library, such as congestion index of each path at each time period, etc., based on historical data.
For example, if the data output by the autonomous lane change model includes lane change left 0.75, lane change right 0.21 and lane change constant 0.04, the current path is adjusted to obtain three corresponding paths: left lane change, right lane change and straight run.
204. And carrying out optimization analysis on the adjusted paths to determine that the optimal path is a lane change path.
Still for example, if the probability of lane change to the left is greater than the probability of lane change to the right and the probability of lane change not according to the lane change decision, the optimal lane change path should be lane change to the left. In this embodiment, however, the multiple paths may also be analyzed to determine the final optimal path.
In a specific example, according to the lane change decision, the probability of lane change to the left is larger than the probability of other lane change decisions, but according to the historical data of each path, the collected roads are arranged at the front 100 m behind the lane change to the left, so that traffic collection is frequently jammed, and the optimal path is finally determined to be unchanged.
As can be seen from the above examples, in this embodiment, the neural network model is first used to determine the lane-changing decision data, determine the rough lane-changing direction, and then the quadratic programming mode is used to determine the specific lane-changing adjustment strategy to obtain the lane-changing path, so that the accurate and reasonable lane-changing path can be obtained.
205. And sending the variable road diameter to a controller of the vehicle simulation model so that the controller generates input control parameters of the vehicle simulation model according to the variable road diameter.
Specifically, the input control parameters include a plurality of kinds, and the present embodiment is described taking an opening angle of an accelerator pedal, a rotation angle of a brake pedal, a rotation direction of a steering wheel, and a rotation angle as an example. The control of the vehicle simulation model lane change running can be realized by controlling the accelerator pedal, the brake pedal and the steering wheel.
206. And controlling the action of a control part of the vehicle simulation model according to the input control parameters so as to control the vehicle simulation model to run along the variable road diameter.
Specifically, step 206 includes:
controlling the action of an accelerator pedal of the vehicle simulation model according to the opening angle of the accelerator pedal;
controlling the action of a brake pedal of the vehicle simulation model according to the rotation angle of the brake pedal; and/or
And controlling the steering wheel of the vehicle simulation model to rotate according to the rotation direction and the rotation angle of the steering wheel.
207. And acquiring current self-state information of the vehicle simulation model through a perception module of the vehicle simulation model, and inputting the current self-state information of the vehicle simulation model into the autonomous lane change model so as to continuously train the autonomous lane change model.
The self state information is used as feedback of the vehicle simulation model and is input into the autonomous lane change model, so that closed-loop training of the autonomous lane change model can be realized, and lane change decision data output by the autonomous lane change model in a subsequent process is more accurate.
According to the automatic lane change planning method provided by the embodiment, the simulation road conforming to the actual scene is established, so that the vehicle simulation model runs in the simulation road to determine the environment information, the success rate of the vehicle simulation model moving from the simulation environment to the real environment is improved, then the environment information is input into the automatic lane change model to obtain lane change decision data, the current path is adjusted based on the lane change decision data to obtain a lane change path, and therefore the lane change decision data is determined by adopting the neural network model firstly, and the rough lane change direction is determined; and then determining a specific lane change adjustment strategy by adopting a quadratic programming mode to obtain a lane change path, thereby obtaining an accurate and reasonable lane change path.
In addition, the embodiment can further input the current self state information into the autonomous lane change model to continuously train the autonomous lane change model, so that final closed loop training is realized.
The automatic lane changing planning device provided by the invention is described below, and the automatic lane changing planning device described below and the automatic lane changing planning method described above can be correspondingly referred to each other.
The embodiment of the invention discloses an automatic lane change planning device which is applied to a vehicle simulation model, and referring to fig. 3, the device comprises:
an environmental information obtaining module 301, configured to control the vehicle simulation model to obtain environmental information in a current path of a simulation road;
the lane change decision generation module 302 is configured to input the environmental information into an autonomous lane change model to obtain lane change decision data;
the lane change path adjustment module 303 is configured to adjust the current path based on lane change decision data to obtain a lane change path;
the lane change driving module 304 is configured to generate control parameters based on a lane change path to control the vehicle simulation model to drive along the lane change path.
Optionally, the environmental information obtaining module 301 is specifically configured to obtain, through a sensing module of the vehicle simulation model, the surrounding vehicle state information and the self state information in the current path of the simulated road; wherein, week car state information includes: obstacle information, relative position, relative velocity, and relative orientation; the self state information includes: engine speed, engine power, gear, vehicle speed, and steering wheel angle.
Optionally, the device further includes a training module, configured to use the environmental sample information as a training sample, and input a lane change decision tag as a training tag to the autonomous lane change model for performing back propagation training, where the lane change decision tag includes a left lane change tag, a non-lane change tag, and a right lane change tag.
Optionally, the lane change decision generation module 302 is specifically configured to: inputting the environment information into an autonomous lane change model to obtain a plurality of lane change decision data and probability values thereof, wherein the lane change decision data comprises left lane change, unchanged lane change and right lane change;
the variable road diameter adjustment module 303 is specifically configured to: according to pre-stored path library information, adjusting the current path based on a plurality of lane change decision data; and carrying out optimization analysis on the adjusted paths to determine that the optimal path is a lane change path.
Optionally, the lane-changing driving module 304 is specifically configured to: the variable road diameter is sent to a controller of a vehicle simulation model, so that the controller generates input control parameters of the vehicle simulation model according to the variable road diameter;
and controlling the action of a control part of the vehicle simulation model according to the input control parameters so as to control the vehicle simulation model to run along the variable road diameter.
Optionally, the input control parameters include: the opening angle of the accelerator pedal, the rotation angle of the brake pedal, the rotation direction and the rotation angle of the steering wheel;
the lane change driving module 304 is specifically configured to:
controlling the action of an accelerator pedal of the vehicle simulation model according to the opening angle of the accelerator pedal;
controlling the action of a brake pedal of the vehicle simulation model according to the rotation angle of the brake pedal; and/or
And controlling the steering wheel of the vehicle simulation model to rotate according to the rotation direction and the rotation angle of the steering wheel.
Optionally, the environmental information obtaining module 301 is specifically configured to: acquiring current self-state information of the vehicle simulation model through a perception module of the vehicle simulation model; and inputting the current self-state information of the vehicle simulation model into the autonomous lane change model so as to continuously train the autonomous lane change model.
According to the automatic lane change planning device provided by the embodiment, the simulation road conforming to the actual scene is established, so that the vehicle simulation model runs in the simulation road to determine the environment information, the success rate of the vehicle simulation model moving from the simulation environment to the real environment is improved, then the environment information is input into the automatic lane change model to obtain lane change decision data, the current path is adjusted based on the lane change decision data to obtain a lane change path, and therefore the lane change decision data is determined by adopting the neural network model firstly, and the rough lane change direction is determined; and then determining a specific lane change adjustment strategy by adopting a quadratic programming mode to obtain a lane change path, thereby obtaining an accurate and reasonable lane change path.
Fig. 4 illustrates a physical schematic diagram of an electronic device, as shown in fig. 4, which may include: processor 410, communication interface (Communications Interface) 420, memory 430 and communication bus 440, wherein processor 410, communication interface 420 and memory 430 communicate with each other via communication bus 440. The processor 410 may invoke logic instructions in the memory 430 to perform an automatic lane change planning method, applied to a vehicle simulation model, comprising:
controlling the vehicle simulation model to acquire environmental information in a current path of a simulation road;
inputting the environmental information into an autonomous lane change model to obtain lane change decision data;
adjusting the current path based on the lane change decision data to obtain a lane change path;
and generating control parameters based on the variable road diameter so as to control the vehicle simulation model to run along the variable road diameter.
Further, the logic instructions in the memory 430 described above may be implemented in the form of software functional units and may be stored in a computer-readable storage medium when sold or used as a stand-alone product. Based on this understanding, the technical solution of the present invention may be embodied essentially or in a part contributing to the prior art or in a part of the technical solution, in the form of a software product stored in a storage medium, comprising several instructions for causing a computer device (which may be a personal computer, a server, a network device, etc.) to perform all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a random access Memory (RAM, random Access Memory), a magnetic disk, or an optical disk, or other various media capable of storing program codes.
In another aspect, the present invention also provides a computer program product comprising a computer program stored on a non-transitory computer readable storage medium, the computer program comprising program instructions which, when executed by a computer, enable the computer to perform the automatic lane change planning method provided by the above methods, applied to a vehicle simulation model, the method comprising:
controlling the vehicle simulation model to acquire environmental information in a current path of a simulation road;
inputting the environmental information into an autonomous lane change model to obtain lane change decision data;
adjusting the current path based on the lane change decision data to obtain a lane change path;
and generating control parameters based on the variable road diameter so as to control the vehicle simulation model to run along the variable road diameter.
In yet another aspect, the present invention further provides a non-transitory computer readable storage medium having stored thereon a computer program which, when executed by a processor, is implemented to perform the above provided automatic lane change planning methods, applied to a vehicle simulation model, the method comprising:
controlling the vehicle simulation model to acquire environmental information in a current path of a simulation road;
inputting the environmental information into an autonomous lane change model to obtain lane change decision data;
adjusting the current path based on the lane change decision data to obtain a lane change path;
and generating control parameters based on the variable road diameter so as to control the vehicle simulation model to run along the variable road diameter.
The apparatus embodiments described above are merely illustrative, wherein the elements illustrated as separate elements may or may not be physically separate, and the elements shown as elements may or may not be physical elements, may be located in one place, or may be distributed over a plurality of network elements. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment. Those of ordinary skill in the art will understand and implement the present invention without undue burden.
From the above description of the embodiments, it will be apparent to those skilled in the art that the embodiments may be implemented by means of software plus necessary general hardware platforms, or of course may be implemented by means of hardware. Based on this understanding, the foregoing technical solution may be embodied essentially or in a part contributing to the prior art in the form of a software product, which may be stored in a computer readable storage medium, such as ROM/RAM, a magnetic disk, an optical disk, etc., including several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute the method described in the respective embodiments or some parts of the embodiments.
Finally, it should be noted that: the above embodiments are only for illustrating the technical solution of the present invention, and are not limiting; although the invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit and scope of the technical solutions of the embodiments of the present invention.
Claims (8)
1. An automatic lane-change planning method, applied to a vehicle simulation model, comprising:
controlling the vehicle simulation model to acquire environmental information in a current path of a simulation road;
inputting the environmental information into an autonomous lane change model to obtain lane change decision data;
adjusting the current path based on the lane change decision data to obtain a lane change path;
generating control parameters based on the variable road diameter so as to control the vehicle simulation model to run along the variable road diameter;
the inputting the environmental information into the autonomous lane change model to obtain lane change decision data includes:
inputting the environment information into an autonomous lane change model to obtain a plurality of lane change decision data and probability values thereof, wherein the lane change decision data comprises left lane change, unchanged lane change and right lane change;
the step of adjusting the current path based on the lane change decision data to obtain a lane change path comprises the following steps:
according to pre-stored path library information, adjusting the current path based on a plurality of lane change decision data;
carrying out optimization analysis on the adjusted paths to determine that the optimal path is a variable path;
the generating control parameters based on the variable road diameter to control the vehicle simulation model to run along the variable road diameter includes:
the variable road diameter is sent to a controller of a vehicle simulation model, so that the controller generates input control parameters of the vehicle simulation model according to the variable road diameter;
and controlling the action of a control part of the vehicle simulation model according to the input control parameters so as to control the vehicle simulation model to run along the variable road diameter, wherein the control part is a physical entity element.
2. The automatic lane-change planning method according to claim 1, wherein controlling the vehicle simulation model to acquire environmental information in a current path of a simulated road comprises:
acquiring the state information of the surrounding vehicles and the state information of the surrounding vehicles in the current path of the simulation road through a perception module of the vehicle simulation model;
wherein, the week car state information includes: obstacle information, relative position, relative velocity, and relative orientation;
the self state information includes: engine speed, engine power, gear, vehicle speed, and steering wheel angle.
3. The automatic lane-change planning method of claim 1 wherein the autonomous lane-change model is trained by:
and taking the environment sample information as a training sample, taking a lane change decision label as a training label, inputting the training label into the autonomous lane change model, and carrying out back propagation training, wherein the lane change decision label comprises a left lane change label, a non-lane change label and a right lane change label.
4. The automatic lane-change planning method of claim 1 wherein the input control parameters comprise: the opening angle of the accelerator pedal, the rotation angle of the brake pedal, the rotation direction and the rotation angle of the steering wheel;
controlling the action of a control component of the vehicle simulation model according to the input control parameters, comprising:
controlling the action of an accelerator pedal of the vehicle simulation model according to the opening angle of the accelerator pedal;
controlling the action of a brake pedal of the vehicle simulation model according to the rotation angle of the brake pedal; and/or
And controlling the steering wheel of the vehicle simulation model to rotate according to the rotation direction and the rotation angle of the steering wheel.
5. The automatic lane-change planning method according to claim 1, further comprising, after controlling the actions of the control components of the vehicle simulation model in accordance with the input control parameters:
acquiring current self-state information of the vehicle simulation model through a perception module of the vehicle simulation model;
and inputting the current self-state information of the vehicle simulation model into the autonomous lane change model so as to continuously train the autonomous lane change model.
6. An automatic lane-changing planning apparatus for use with a vehicle simulation model, the apparatus comprising:
the environment information acquisition module is used for controlling the vehicle simulation model to acquire environment information in the current path of the simulation road;
the lane change decision generation module is used for inputting the environmental information into an autonomous lane change model to obtain lane change decision data;
the lane change path adjusting module is used for adjusting the current path based on lane change decision data to obtain a lane change path;
the lane change driving module is used for generating control parameters based on a lane change path so as to control the vehicle simulation model to drive along the lane change path;
the lane change decision generation module is specifically configured to: inputting the environment information into an autonomous lane change model to obtain a plurality of lane change decision data and probability values thereof, wherein the lane change decision data comprises left lane change, unchanged lane change and right lane change;
the variable road diameter adjusting module is specifically used for: according to pre-stored path library information, adjusting the current path based on a plurality of lane change decision data; carrying out optimization analysis on the adjusted paths to determine that the optimal path is a variable path;
the lane change driving module is specifically configured to send the lane change path to a controller of a vehicle simulation model, so that the controller generates input control parameters of the vehicle simulation model according to the lane change path; and controlling the action of a control part of the vehicle simulation model according to the input control parameters so as to control the vehicle simulation model to run along the variable road diameter, wherein the control part is a physical entity element.
7. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the processor implements the steps of the automatic lane change planning method of any one of claims 1 to 5 when the program is executed by the processor.
8. A non-transitory computer readable storage medium having stored thereon a computer program, characterized in that the computer program when executed by a processor implements the steps of the automatic lane change planning method of any of claims 1 to 5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011593899.4A CN112835362B (en) | 2020-12-29 | 2020-12-29 | Automatic lane change planning method and device, electronic equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011593899.4A CN112835362B (en) | 2020-12-29 | 2020-12-29 | Automatic lane change planning method and device, electronic equipment and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112835362A CN112835362A (en) | 2021-05-25 |
| CN112835362B true CN112835362B (en) | 2023-06-30 |
Family
ID=75925088
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011593899.4A Active CN112835362B (en) | 2020-12-29 | 2020-12-29 | Automatic lane change planning method and device, electronic equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112835362B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113324556B (en) * | 2021-06-04 | 2024-03-26 | 苏州智加科技有限公司 | Path planning method and device based on vehicle-road collaborative reinforcement learning and application system |
| CN114906142A (en) * | 2022-04-29 | 2022-08-16 | 中汽创智科技有限公司 | Vehicle cooperative lane changing method and device, electronic equipment and storage medium |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2007322183A (en) * | 2006-05-31 | 2007-12-13 | Sanyo Electric Co Ltd | In-vehicle navigation apparatus |
Family Cites Families (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP4822099B2 (en) * | 2005-07-11 | 2011-11-24 | アイシン・エィ・ダブリュ株式会社 | Navigation device and navigation method |
| CN105955253A (en) * | 2016-04-20 | 2016-09-21 | 郑州宇通客车股份有限公司 | Method and system for controlling turn lights and vehicle doors of autonomous driving vehicle |
| CN107991898A (en) * | 2016-10-26 | 2018-05-04 | 法乐第(北京)网络科技有限公司 | A kind of automatic driving vehicle simulating test device and electronic equipment |
| CN107506830A (en) * | 2017-06-20 | 2017-12-22 | 同济大学 | Towards the artificial intelligence training platform of intelligent automobile programmed decision-making module |
| WO2018235159A1 (en) * | 2017-06-20 | 2018-12-27 | 三菱電機株式会社 | Path estimation device and path estimation method |
| US11130497B2 (en) * | 2017-12-18 | 2021-09-28 | Plusai Limited | Method and system for ensemble vehicle control prediction in autonomous driving vehicles |
| WO2019127079A1 (en) * | 2017-12-27 | 2019-07-04 | Bayerische Motoren Werke Aktiengesellschaft | Vehicle lane change prediction |
| JP7085859B2 (en) * | 2018-02-23 | 2022-06-17 | 本田技研工業株式会社 | Vehicle control unit |
| EP3667556A1 (en) * | 2018-12-12 | 2020-06-17 | Visteon Global Technologies, Inc. | Autonomous lane change |
| CN111665738A (en) * | 2019-03-05 | 2020-09-15 | 阿里巴巴集团控股有限公司 | In-loop simulation system and information processing method and device thereof |
| KR20190083317A (en) * | 2019-06-20 | 2019-07-11 | 엘지전자 주식회사 | An artificial intelligence apparatus for providing notification related to lane-change of vehicle and method for the same |
| CN110379160A (en) * | 2019-07-09 | 2019-10-25 | 广州小鹏汽车科技有限公司 | A kind of automatic lane change householder method of the vehicle of traffic intersection, system and vehicle |
| CN110435658A (en) * | 2019-07-19 | 2019-11-12 | 中国第一汽车股份有限公司 | A kind of control method for vehicle, device, vehicle and storage medium |
| CN110949391B (en) * | 2019-11-04 | 2021-07-13 | 广州小鹏汽车科技有限公司 | Vehicle lane change control method and system, vehicle and storage medium |
| CN111775938B (en) * | 2020-06-24 | 2021-11-16 | 福瑞泰克智能***有限公司 | Lane change path planning method, device and system |
-
2020
- 2020-12-29 CN CN202011593899.4A patent/CN112835362B/en active Active
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2007322183A (en) * | 2006-05-31 | 2007-12-13 | Sanyo Electric Co Ltd | In-vehicle navigation apparatus |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112835362A (en) | 2021-05-25 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110969848B (en) | Automatic driving overtaking decision method based on reinforcement learning under opposite double lanes | |
| US20240140487A1 (en) | Autonomous Vehicles Featuring Machine-Learned Yield Model | |
| JP2022532972A (en) | Unmanned vehicle lane change decision method and system based on hostile imitation learning | |
| CN107168303A (en) | A kind of automatic Pilot method and device of automobile | |
| CN112835362B (en) | Automatic lane change planning method and device, electronic equipment and storage medium | |
| US20210341886A1 (en) | System and Method of Efficient, Continuous, and Safe Learning Using First Principles and Constraints | |
| CN116134292A (en) | Tool for performance testing and/or training an autonomous vehicle planner | |
| CN111661045B (en) | Training generator unit and discriminator unit for collision-aware trajectory prediction | |
| Fehér et al. | Hierarchical evasive path planning using reinforcement learning and model predictive control | |
| Chen et al. | ES-DQN: A learning method for vehicle intelligent speed control strategy under uncertain cut-in scenario | |
| Wang et al. | High-level decision making for automated highway driving via behavior cloning | |
| Yan et al. | A multi-vehicle game-theoretic framework for decision making and planning of autonomous vehicles in mixed traffic | |
| Yu et al. | A Novel Dynamic Lane‐Changing Trajectory Planning Model for Automated Vehicles Based on Reinforcement Learning | |
| Cai et al. | Rule‐constrained reinforcement learning control for autonomous vehicle left turn at unsignalized intersection | |
| Wei et al. | A learning-based autonomous driver: emulate human driver's intelligence in low-speed car following | |
| Wang et al. | A deep reinforcement learning-based approach for autonomous lane-changing velocity control in mixed flow of vehicle group level | |
| Zhu et al. | Design of an integrated vehicle chassis control system with driver behavior identification | |
| Guo et al. | Optimal design of a driver assistance controller based on surrounding vehicle’s social behavior game model | |
| Huang et al. | Human Knowledge Enhanced Reinforcement Learning for Mandatory Lane-Change of Autonomous Vehicles in Congested Traffic | |
| Yang et al. | Decision-making in autonomous driving by reinforcement learning combined with planning & control | |
| CN114987511A (en) | Method for simulating human driving behavior to train neural network-based motion controller | |
| Wang et al. | An end-to-end deep reinforcement learning model based on proximal policy optimization algorithm for autonomous driving of off-road vehicle | |
| Gutiérrez-Moreno et al. | Hybrid decision making for autonomous driving in complex urban scenarios | |
| Yang et al. | Deep Reinforcement Learning Lane-Changing Decision Algorithm for Intelligent Vehicles Combining LSTM Trajectory Prediction | |
| Zhao et al. | Real-time Learning of Driving Gap Preference for Personalized Adaptive Cruise Control |
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 |