CN116499487B - Vehicle path planning method, device, equipment and medium - Google Patents
Vehicle path planning method, device, equipment and medium Download PDFInfo
- Publication number
- CN116499487B CN116499487B CN202310770396.7A CN202310770396A CN116499487B CN 116499487 B CN116499487 B CN 116499487B CN 202310770396 A CN202310770396 A CN 202310770396A CN 116499487 B CN116499487 B CN 116499487B
- Authority
- CN
- China
- Prior art keywords
- vehicle
- historical
- moment
- target
- state information
- 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
- 238000000034 method Methods 0.000 title claims abstract description 59
- 238000012549 training Methods 0.000 claims abstract description 10
- 239000013598 vector Substances 0.000 claims description 77
- 230000006870 function Effects 0.000 claims description 63
- 238000004590 computer program Methods 0.000 claims description 16
- 238000012545 processing Methods 0.000 claims description 16
- 230000004888 barrier function Effects 0.000 claims description 15
- 230000008569 process Effects 0.000 claims description 14
- 230000009471 action Effects 0.000 abstract description 5
- 230000008859 change Effects 0.000 abstract description 4
- 230000001133 acceleration Effects 0.000 description 10
- 238000004891 communication Methods 0.000 description 8
- 238000010586 diagram Methods 0.000 description 7
- 239000003795 chemical substances by application Substances 0.000 description 4
- 230000003287 optical effect Effects 0.000 description 3
- 230000003993 interaction Effects 0.000 description 2
- 230000007774 longterm Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000005457 optimization Methods 0.000 description 2
- 230000003068 static effect Effects 0.000 description 2
- 238000003491 array Methods 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000001413 cellular effect Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 239000011521 glass Substances 0.000 description 1
- 239000004973 liquid crystal related substance Substances 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000007726 management method Methods 0.000 description 1
- 239000013307 optical fiber Substances 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 230000001953 sensory effect Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
- G01C21/00—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00
- G01C21/26—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00 specially adapted for navigation in a road network
- G01C21/34—Route searching; Route guidance
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
- G01C21/00—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00
- G01C21/26—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00 specially adapted for navigation in a road network
- G01C21/34—Route searching; Route guidance
- G01C21/3446—Details of route searching algorithms, e.g. Dijkstra, A*, arc-flags, using precalculated routes
-
- 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)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Automation & Control Theory (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Traffic Control Systems (AREA)
Abstract
The invention discloses a vehicle path planning method, device, equipment and medium. Wherein the method comprises the following steps: determining target data of a target vehicle at the current moment; obtaining the own vehicle state information of the target vehicle at least one first prediction moment and the own vehicle state information of the target vehicle at a second prediction moment according to the target data and based on the prediction model; the first predicted time is earlier than the second predicted time; the prediction model is obtained by training based on historical data of at least one historical vehicle; and determining a target running path of the target vehicle according to the target data, the vehicle state information at the first prediction time and the vehicle state information at the second prediction time. By executing the scheme, the similarity between the track planning result and the manual driving estimation can be improved, the track stability can be improved, the continuity of the action space is ensured, and the problem of controlling the dragon drawing caused by frequent track change is solved.
Description
Technical Field
The present invention relates to the field of path planning technologies for automatically driving vehicles, and in particular, to a method, an apparatus, a device, and a medium for path planning of a vehicle.
Background
The automatic driving vehicle is an intelligent automobile which senses the road environment through a vehicle-mounted sensing system, automatically plans a driving route and controls the vehicle to reach a preset target. For an autonomous vehicle, path planning is an important factor in relation to the safety and stability of the vehicle's travel.
In the related art, the vehicle driving track prediction scheme has limited factors influencing driving considered in the track planning process, and limited set constraint conditions, and only predicts the track in a certain future time (usually, the time is shorter, such as 1 second). The problem of 'dragon drawing' occurs in the running process of the vehicle, such as the left-right swing of the vehicle, and the unstable speed and acceleration of the vehicle in the running process along the predicted track increase the risk of road running.
Disclosure of Invention
The invention provides a vehicle path planning method, device, equipment and medium, which can improve the similarity between a track planning result and manual driving estimation, improve track stability, ensure continuity of an action space and reduce the problem of controlling the picture dragon caused by frequent change of the track.
According to an aspect of the present invention, there is provided a path planning method of a vehicle, the method comprising:
Determining target data of a target vehicle at the current moment; the target data comprises vehicle body information, vehicle state information, barrier information and map information; the vehicle state information comprises position information and motion state information;
obtaining the own vehicle state information of the target vehicle at least one first prediction moment and the own vehicle state information of the target vehicle at a second prediction moment according to the target data and based on a prediction model; the first predicted time is earlier than the second predicted time; the prediction model is obtained by training based on historical data of at least one historical vehicle;
and determining a target running path of the target vehicle according to the target data, the self-vehicle state information at the first prediction time and the self-vehicle state information at the second prediction time.
According to another aspect of the present invention, there is provided a path planning apparatus of a vehicle, the apparatus comprising:
the target data determining module is used for determining target data of the target vehicle at the current moment; the target data comprises vehicle body information, vehicle state information, barrier information and map information; the vehicle state information comprises position information and motion state information;
The future time own vehicle state information determining module is used for obtaining own vehicle state information of the target vehicle at least one first prediction time and own vehicle state information of the target vehicle at a second prediction time according to the target data and based on a prediction model; the first predicted time is earlier than the second predicted time; the prediction model is obtained by training based on historical data of at least one historical vehicle;
and the target driving path determining module is used for determining a target driving path of the target vehicle according to the target data, the vehicle state information at the first prediction moment and the vehicle state information at the second prediction moment.
According to another aspect of the present invention, there is provided an electronic apparatus including:
at least one processor; and
a memory communicatively coupled to the at least one processor; wherein,
the memory stores a computer program executable by the at least one processor to enable the at least one processor to perform the method of path planning for a vehicle according to any one of the embodiments of the present invention.
According to another aspect of the present invention, there is provided a computer readable storage medium storing computer instructions for causing a processor to execute a path planning method of a vehicle according to any one of the embodiments of the present invention.
According to the technical scheme, the target data of the target vehicle at the current moment is determined; the target data includes vehicle body information, vehicle state information, obstacle information, and map information; the vehicle state information includes position information and motion state information; obtaining the own vehicle state information of the target vehicle at least one first prediction moment and the own vehicle state information of the target vehicle at a second prediction moment according to the target data and based on the prediction model; the first predicted time is earlier than the second predicted time; the prediction model is obtained by training based on historical data of at least one historical vehicle; and determining a target running path of the target vehicle according to the target data, the vehicle state information at the first prediction time and the vehicle state information at the second prediction time. By executing the scheme provided by the embodiment of the invention, the similarity between the track planning result and the manual driving estimation can be improved, the track stability can be improved, the continuity of the action space is ensured, and the problem of controlling the picture dragon caused by frequent change of the track is reduced.
It should be understood that the description in this section is not intended to identify key or critical features of the embodiments of the invention or to delineate the scope of the invention. Other features of the present invention will become apparent from the description that follows.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for the description of the embodiments will be briefly described below, and it is apparent that the drawings in the following description are only some embodiments of the present invention, and other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
Fig. 1 is a flowchart of a path planning method for a vehicle according to an embodiment of the present invention;
FIG. 2 is a flow chart of another vehicle path planning method provided by an embodiment of the present invention;
FIG. 3 is a flow chart of a determination process of a predictive model provided by an embodiment of the invention;
FIG. 4 is a schematic diagram of a second real track time point in the historical data provided by the embodiment of the invention;
FIG. 5 is a schematic diagram of a target motion path of a target vehicle according to an embodiment of the present invention;
fig. 6 is a schematic structural diagram of a path planning apparatus for a vehicle according to an embodiment of the present invention;
Fig. 7 is a schematic structural diagram of an electronic device implementing a path planning method of a vehicle according to an embodiment of the present invention.
Detailed Description
In order that those skilled in the art will better understand the present invention, a technical solution in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in which it is apparent that the described embodiments are only 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 present invention without making any inventive effort, shall fall within the scope of the present invention.
It should be noted that the terms "first," "second," and the like in the description and the claims of the present invention and the above figures are used for distinguishing between similar objects and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used may be interchanged where appropriate such that the embodiments of the invention described herein may be implemented in sequences other than those illustrated or otherwise described herein. Furthermore, the terms "comprises," "comprising," and "having," and any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus.
Fig. 1 is a flowchart of a vehicle path planning method provided in an embodiment of the present invention, where the present embodiment is applicable to a case of planning a driving path of an autonomous vehicle, the method may be performed by a vehicle path planning apparatus, and the vehicle path planning apparatus may be implemented in hardware and/or software, and the vehicle path planning apparatus may be configured in an electronic device for path planning of a vehicle. As shown in fig. 1, the method includes:
and S110, determining target data of the target vehicle at the current moment.
Wherein the target data includes vehicle body information, vehicle state information, obstacle information, and map information; the own vehicle state information includes position information and motion state information.
The target vehicle is an own vehicle, and the present solution may use a Frenet coordinate system, and determine target data of the target vehicle at the current moment by using a position of the target vehicle at the current moment as an origin of coordinates. The body information may be a length and a width of the target vehicle. The obstacle information may be information of dynamic obstacles and static obstacles around the target vehicle at the current moment, the dynamic obstacles may be vehicles and pedestrians, and the static obstacles may be roadblocks and road edges. Taking an obstacle as an example of a vehicle, the obstacle information may be at least one of a longitudinal coordinate, a transverse coordinate, a head orientation, a longitudinal speed, a transverse speed, a vehicle length, and a vehicle width of the obstacle. The position information may be at least one of longitudinal coordinates, lateral coordinates, and head orientation of the target vehicle. The movement state information may be at least one of a longitudinal speed, a lateral speed, a longitudinal acceleration, and a lateral acceleration of the target vehicle. The map information is high definition map data.
And S120, obtaining the own vehicle state information of the target vehicle at least one first prediction moment and the own vehicle state information of the target vehicle at a second prediction moment according to the target data and the prediction model.
Wherein the first predicted time is earlier than the second predicted time; the predictive model is trained based on historical data of at least one historical vehicle.
For example, the historical data may include time information, movement state information, and position information of the historical vehicle at each track point. According to the scheme, the reference line of the lane where the target vehicle is located at the current moment can be determined according to the map information at the current moment and the vehicle state information of the target vehicle at the current moment, then 100 points closest to the coordinate origin are determined from the reference line at intervals of 20cm according to the position of the target vehicle at the current moment to serve as reference points, 70 reference points are located in front of the coordinate origin, and 30 reference points are located behind the coordinate origin. And determining information of each reference point of the target vehicle at the current moment, wherein the information of the reference point can be represented by longitudinal coordinates of the reference point, transverse coordinates of the reference point, length of a reference line, orientation of the reference line and curvature. The reference line length is the distance between the reference point and the origin of coordinates. The reference line orientation may be determined from map information at the current time. The curvature may be the curvature of a straight line connecting the current reference point with the previous reference point. In the scheme, the vehicle body information and the vehicle state information in the target data are represented by A, the obstacle information is represented by B, the reference point information of the target vehicle at the current moment is represented by C, and the vehicle body information in the target data is the vehicle length and the vehicle width of the target vehicle. The vehicle state information in the target data comprises longitudinal coordinates, transverse coordinates, head orientation, longitudinal speed, transverse speed, longitudinal acceleration and transverse acceleration of the target vehicle. The number of the barriers can be set according to actual needs, for example, 64, and the barrier information comprises a longitudinal coordinate, a transverse coordinate, a vehicle head direction, a longitudinal speed, a transverse speed, a vehicle length and a vehicle width. The number of the reference points is 100, and the reference point information comprises a reference point longitudinal coordinate, a reference point transverse coordinate, a reference line length, a reference line orientation and curvature. The input vector dimensions of A, B and C are different, the vector dimension of A is [1,9], B is [64,7], and C is [100,5], and the scheme can respectively carry out vectorization processing on A, B and C to obtain vectors A ', B' and C ', wherein the vectorization scale is 128, namely the vector dimension of A' is [1, 128], the vector dimension of B 'is [64, 128], and the vector dimension of C' is [100, 128].
Then, the a ', B' and C 'determined in the previous step are spliced to obtain a vector AA (k vector) and a vector BB (v vector), and the vector a' is taken as a q vector. Vector dimensions for vector AA and vector BB are both 165, 128. The vectors AA, a', and BB are input into the prediction model to obtain the own vehicle state information of the target vehicle at least one first prediction time and the own vehicle state information of the target vehicle at a second prediction time, as shown in fig. 2. The first predicted time is between the current time and the second predicted time. The method can realize the constraint of long-term targets and short-term targets on the driving path of the vehicle in the track planning process, and can realize the driving stability of the vehicle and optimize the driving track compared with the track planning scheme of only focusing on the short-term targets in the prior art.
In this embodiment, optionally, the determining process of the prediction model includes: for each historical data, determining starting data of the historical vehicle according to the historical data; the initial data comprise vehicle body information, vehicle state information, barrier information and reference point information of the historical vehicle at the initial moment in the historical data; the reference point is determined according to the initial data and a preset distance; inputting the initial data into a preset prediction algorithm to obtain the vehicle state information of the first historical prediction moment and the vehicle state information of the second historical prediction moment; determining the vehicle state information of the historical vehicle at the first real track moment in the historical data; the first real track moment is determined according to the starting moment and a preset period; determining the own vehicle state information of the historical vehicle at the second real track moment in the historical data; and determining the prediction model according to the initial data, the own vehicle state information of each historical prediction moment historical vehicle, the own vehicle state information of each actual track moment historical vehicle and a loss function of a preset prediction algorithm.
As shown in fig. 3, taking the history data of a certain history vehicle as an example, the present solution may use a Frenet coordinate system, and determine the starting data of the history vehicle at the starting time by using the position of the history vehicle at the starting time in the history data as the origin of coordinates. The vehicle body information in the initial data is the vehicle length and the vehicle width of the historical vehicle. The own vehicle state information in the start data includes position information and motion state information of the history vehicle at the start time. The location information may be at least one of longitudinal coordinates, lateral coordinates, and head orientation of the historical vehicle. The movement state information may be at least one of a longitudinal speed, a lateral speed, a longitudinal acceleration, a lateral acceleration of the history vehicle.
The obstacle information in the start data may be information of 64 obstacles closest to the position of the history vehicle at the start time, and the obstacle information may be represented by a longitudinal coordinate, a transverse coordinate, a head orientation, a longitudinal speed, a transverse speed, a length of the vehicle, and a width of the vehicle.
The preset distance can be set according to actual needs, for example, 20cm. According to the scheme, the reference line of a lane where the historical vehicle is located can be determined according to the position information of the historical vehicle at the starting moment, a plurality of reference points on the reference line are further determined according to the preset distance and the position information of the historical vehicle at the starting moment, then the reference point information of each reference point is determined according to the map data at the starting moment, and the reference point information is represented by the longitudinal coordinate of the reference point, the transverse coordinate of the reference point, the length of the reference line, the orientation of the reference line and the curvature.
The preset prediction algorithm can be selected according to actual needs, and the initial data of the historical vehicles are processed through the preset prediction algorithm to obtain the self-vehicle state information of at least one historical vehicle at the first historical prediction moment and the self-vehicle state information of the historical vehicle at the second historical prediction moment, wherein the first historical prediction moment is earlier than the second historical prediction moment.
The preset period may be set according to actual needs, for example, the preset period may be 2s. The scheme can be used for determining at least one first real track moment from the starting moment according to a preset period and determining the own vehicle state information of the historical vehicle at the first real track moment in the historical data. The scheme can also determine the second real track moment in the historical data, and further determine the own vehicle state information of the historical vehicle at the second real track moment in the historical data. The own vehicle state information may be position information and motion state information. The second real track time may be a time 60 frames apart from the start time, or the position information of the second real track time history vehicle deviates from the reference line of the lane in which the history vehicle is located. The interval time between two adjacent track points in the historical data can be 0.1s.
Finally, the scheme can train a preset prediction algorithm based on a preset loss function according to the initial data determined by each historical data, the own vehicle state information of each historical prediction time historical vehicle and the own vehicle state information of each real track time historical vehicle so as to determine a prediction model. For example, when the loss value of the loss function converges or the loss value of the loss function meets a certain condition, the parameter optimization in the preset prediction algorithm is completed, and the preset prediction algorithm with the completed parameter optimization is used as a prediction model. The prediction model can be determined through the historical data of the historical vehicle, and a reliable data base is provided for subsequent path planning.
In a possible embodiment, optionally, the determining of the reference point includes: determining a target lane where the historical vehicle is located according to the historical data; determining a target reference line of the target lane; and determining a preset number of reference points from the target reference line according to the initial data and a preset distance.
Taking a preset distance of 20cm as an example, the scheme can determine a reference line of a lane where a historical vehicle is located according to map data of a starting moment and position information of the historical vehicle at the starting moment, then determine 100 points closest to a coordinate origin from the reference line every 20cm based on the position information of the historical vehicle at the starting moment as reference points, wherein 70 reference points are located in front of the coordinate origin, and 30 reference points are located behind the coordinate origin. Can be implemented to provide reliable training data for deriving a predictive model.
In another possible embodiment, optionally, the determining of the second real track moment includes: taking the moment of a preset time period from the starting moment in the historical data as the moment of a second real track; or determining the orientation angle of each track point in the historical data, determining the orientation angle difference value between the orientation angle of each track point and the orientation angle of the previous track point, and determining the moment of the track point in the historical data of the historical vehicle as the second real track moment when the absolute value of the orientation angle difference value is larger than a preset threshold value.
The second real track time may be determined from the historical data based on the starting time, for example, as shown in fig. 4, where the second real track time may be a time corresponding to a node (black waypoint) where a jump occurs in the historical data. According to the scheme, the orientation angle of each track point in the historical data can be obtained, the orientation angle difference value of the heading angle of each track point and the orientation angle of the previous track point is determined, and when the absolute value of the orientation angle difference value is larger than a preset threshold value, the moment of the track point of the historical vehicle in the historical data is determined to be the second real track moment. In addition, if the node where the jump occurs is not found up to the 60 th frame (each frame is 0.1 s) by the above scheme for determining the second real track time, and the longitudinal distance between the position corresponding to the 60 th frame in the historical data and the position of the historical vehicle at the starting time exceeds 2.0 meters, the 60 th frame in the historical data is selected as the second real track time. The method can realize the determination of a long-term constraint target and provide reliable constraint conditions for determining a better planning track.
In yet another possible embodiment, optionally, the preset predictive algorithm includes a self-attention algorithm; the network structure of the self-attention algorithm comprises a self-attention layer and a multi-head-multi-layer full-connection layer; the determining process of the self-vehicle state information at the first historical predicting moment and the self-vehicle state information at the second historical predicting moment comprises the following steps: vectorizing the vehicle body information and the vehicle state information in the initial data to obtain a first vector; carrying out vectorization processing on the barrier information in the initial data to obtain a second vector; vectorizing the reference point information in the initial data to obtain a third vector; splicing the first vector, the second vector and the third vector to obtain a first target vector and a second target vector; processing the first target vector, the first vector and the second target vector through the self-attention layer and the multi-head-multi-layer full-connection layer to obtain a target planning result of the historical vehicle at the first historical prediction moment and a target planning result of the historical vehicle at the second historical prediction moment; the target planning result comprises a space planning result and a motion planning result.
In this scheme, the vehicle body information and the vehicle state information in the initial data may be represented by ego, the obstacle information may be represented by agent, the reference point information may be represented by refline, and the vehicle body information in the initial data may be the vehicle length and the vehicle width of the historical vehicle. The vehicle state information in the initial data comprises longitudinal coordinates, transverse coordinates, head orientation, longitudinal speed, transverse speed, longitudinal acceleration and transverse acceleration of the historical vehicle. The number of the barriers is 64, and the barrier information comprises a longitudinal coordinate, a transverse coordinate, a headstock direction, a longitudinal speed, a transverse speed, a vehicle length and a vehicle width. The number of the reference points is 100, and the reference point information comprises a reference point longitudinal coordinate, a reference point transverse coordinate, a reference line length, a reference line orientation and curvature. ego, agents, refline have different input vector dimensions, the vector dimension of ego is [1,9], the agent is [64,7], and refline is [100,5], and according to the scheme, ego, agents, refline can be subjected to vectorization processing to obtain a first vector, a second vector and a third vector, the vectorization dimensions are all 128, namely the vector dimension of the first vector is [1, 128], the vector dimension of the second vector is [64, 128], and the vector dimension of the third vector is [100, 128].
The first vector, the second vector and the third vector determined in the previous step can be spliced to obtain a first target vector (k vector) and a second target vector (v vector), and the first vector is taken as a q vector. The vector dimensions of both the first and second target vectors are [165, 128]. And taking the first target vector, the first vector and the second target vector as the input of a self-attention layer in the self-attention algorithm to obtain data processed by the self-attention layer, and taking the data processed by the self-attention layer as the input of a multi-head multi-layer full-connection layer in the self-attention algorithm to obtain at least one target planning result of the historical vehicle at the first historical prediction moment and one target planning result of the historical vehicle at the second historical prediction moment. The space planning result in the target planning result can be position information, and the motion planning result can be motion state information. Can be implemented to provide reliable training data for deriving a predictive model.
In yet another possible embodiment, the loss function is optionally determined based on the following formula:
；
wherein ,representing the loss value of the loss function, wherein alpha represents the overall loss function weight of the target at the moment of the first real track, and beta represents the overall loss function weight of the target at the moment of the second real track; Respectively representing a first real track moment space target loss function, a first real track moment moving target loss function, a second real track moment space target loss function and a second real track moment moving target loss function; γ1, γ2, γ3, γ4 respectively represent the first real track moment spatial target loss function weight, the first real track moment moving target loss function weight, the second real track moment spatial target loss function weight, and the second real track moment moving target loss function weight.
For example, α is the overall loss function weight of the target at the moment of the first real track, and the value may be set according to actual needs, for example, 0.7. Beta is the overall loss function weight of the target at the moment of the second real track, and the value can be set according to actual needs, for example, 0.3. The values of γ1, γ2, γ3 and γ4 can be set according to actual needs, and can be the same or different, for example, the values are all 0.5. The spatial target may include longitudinal coordinates, lateral coordinates, and head orientation of the historical vehicle. The moving object may include a speed, an acceleration of the historical vehicle. The method can achieve a better planning path and provide a reliable data base for the running stability of the vehicle.
In yet another possible embodiment, optionally, the first real track moment spatial target loss function includes a difference between the position information of the historical vehicle at the first historical predicted moment and the position information of the historical vehicle at the first real track moment; the first real track moment moving target loss function comprises a difference between the moving state information of the historical vehicle at a first historical predicting moment and the moving state information of the historical vehicle at the first real track moment; the second real track moment space target loss function comprises the difference between the position information of the historical vehicle at the second historical prediction moment and the position information of the historical vehicle at the second real track moment; the second real track moment moving target loss function comprises a difference between the moving state information of the historical vehicle at the second historical predicting moment and the moving state information of the historical vehicle at the second real track moment. The difference between the information of the historical vehicle at the first historical predicted time and the information of the historical vehicle at the first actual track time can be set according to actual needs, for example, the difference between the information can be the difference. Since the position information of the first historical prediction time predicted by the self-attention algorithm is not consistent with the position information of the first real track time in the historical data in a large probability, the motion state information of the first historical prediction time predicted by the self-attention algorithm is not consistent with the motion state of the first real track time in the historical data, the position information of the second historical prediction time predicted by the self-attention algorithm is not consistent with the position information of the second real track time in the historical data, and the motion state information of the second historical prediction time predicted by the self-attention algorithm is not consistent with the motion state information of the second real track time in the historical data. The method may set the first real track moment spatial target loss function to include a difference between the position information of the history vehicle at the first history prediction moment and the position information of the history vehicle at the first real track moment. The first real-track-time moving object loss function is set to include a difference between the moving state information of the history vehicle at the first history prediction time and the moving state information of the history vehicle at the first real-track time. The second real track moment spatial target loss function is set to include a difference between the position information of the history vehicle at the second history prediction moment and the position information of the history vehicle at the second real track moment. The second real trajectory time moving object loss function is set to include a difference between the moving state information of the history vehicle at the second history prediction time and the moving state information of the history vehicle at the second real trajectory time. The method can be used for obtaining a better planning path and providing a reliable data base for the running stability of the vehicle. The method can be used for obtaining a better planning path and providing a reliable data base for the running stability of the vehicle.
S130, determining a target running path of the target vehicle according to the target data, the self-vehicle state information at the first prediction time and the self-vehicle state information at the second prediction time.
As shown in fig. 2, the present solution may process the target data, the vehicle state information at each first predicted time and the vehicle state information at each second predicted time by using an optimal trajectory algorithm and a fifth order polynomial to obtain the vehicle state information of the target vehicle at each time between the current time and the first predicted time, each time between each first predicted time and each time between the first predicted time and the second predicted time, that is, determine the vehicle state information at each time continuous time within a future period, and may obtain the target travel path of the target vehicle, so as to complete the travel trajectory planning of the target vehicle, as shown in fig. 5.
According to the technical scheme, the target data of the target vehicle at the current moment is determined; the target data includes vehicle body information, vehicle state information, obstacle information, and map information; the vehicle state information includes position information and motion state information; obtaining the own vehicle state information of the target vehicle at least one first prediction moment and the own vehicle state information of the target vehicle at a second prediction moment according to the target data and based on the prediction model; the first predicted time is earlier than the second predicted time; the prediction model is obtained by training based on historical data of at least one historical vehicle; and determining a target running path of the target vehicle according to the target data, the vehicle state information at the first prediction time and the vehicle state information at the second prediction time. By executing the scheme provided by the embodiment of the invention, the similarity between the track planning result and the manual driving estimation can be improved, the track stability can be improved, the continuity of the action space is ensured, and the problem of controlling the picture dragon caused by frequent change of the track is reduced.
Fig. 6 is a schematic structural diagram of a path planning apparatus for a vehicle according to an embodiment of the present invention. As shown in fig. 6, the apparatus includes:
a target data determining module 310, configured to determine target data of a target vehicle at a current moment; the target data comprises vehicle body information, vehicle state information, barrier information and map information; the vehicle state information comprises position information and motion state information;
a future time own vehicle state information determining module 320, configured to obtain, according to the target data and based on a prediction model, own vehicle state information of the target vehicle at least one first predicted time and own vehicle state information of the target vehicle at a second predicted time; the first predicted time is earlier than the second predicted time; the prediction model is obtained by training based on historical data of at least one historical vehicle;
the target driving path determining module 330 is configured to determine a target driving path of the target vehicle according to the target data, the vehicle state information at the first predicted time, and the vehicle state information at the second predicted time.
Optionally, the device further includes a prediction model determining module, including a start data determining unit, configured to determine, for each historical data, start data of the historical vehicle according to the historical data before obtaining, according to the target data and based on a prediction model, vehicle state information of the target vehicle at least one first prediction time and vehicle state information of the target vehicle at a second prediction time; the initial data comprise vehicle body information, vehicle state information, barrier information and reference point information of the historical vehicle at the initial moment in the historical data; the reference point is determined according to the initial data and a preset distance; the historical prediction moment self-vehicle state information determining unit is used for inputting the initial data into a preset prediction algorithm to obtain self-vehicle state information of the first historical prediction moment and self-vehicle state information of the second historical prediction moment; a first vehicle state information determining unit of a historical vehicle is used for determining vehicle state information of the historical vehicle at a first real track moment in the historical data; the first real track moment is determined according to the starting moment and a preset period; a second vehicle state information determining unit of the historical vehicle, configured to determine vehicle state information of the historical vehicle at a second real track moment in the historical data; and the prediction model determining unit is used for determining the prediction model according to the initial data, the own vehicle state information of each historical prediction moment historical vehicle, the own vehicle state information of each actual track moment historical vehicle and a loss function of a preset prediction algorithm.
Optionally, the initial data determining unit is specifically configured to determine, according to the history data, a target lane in which the history vehicle is located; determining a target reference line of the target lane; and determining a preset number of reference points from the target reference line according to the initial data and a preset distance.
Optionally, the second vehicle state information determining unit of the historical vehicle is specifically configured to use a time of a preset time period from a starting time in the historical data as a second real track time; or determining the orientation angle of each track point in the historical data in turn, determining the orientation angle difference value between the orientation angle of each track point and the orientation angle of the previous track point, and determining the moment of the track point in the historical data of the historical vehicle as the second real track moment when the absolute value of the orientation angle difference value is larger than a preset threshold value.
Optionally, the preset prediction algorithm includes a self-attention algorithm; the network structure of the self-attention algorithm comprises a self-attention layer and a multi-head-multi-layer full-connection layer; the vehicle state information determining unit at the historical prediction moment comprises a first vector determining subunit, and is used for carrying out vectorization processing on the vehicle body information and the vehicle state information in the initial data to obtain a first vector; a second vector determination subunit, configured to perform vectorization processing on the obstacle information in the initial data to obtain a second vector; a third vector determination subunit, configured to perform vectorization processing on the reference point information in the initial data to obtain a third vector; the splicing subunit is used for splicing the first vector, the second vector and the third vector to obtain a first target vector and a second target vector; the target planning result determining subunit is used for processing the first target vector, the first vector and the second target vector through the self-attention layer and the multi-head-multi-layer full-connection layer to obtain a target planning result of the historical vehicle at the first historical prediction moment and a target planning result of the historical vehicle at the second historical prediction moment; the target planning result comprises a space planning result and a motion planning result.
Optionally, the prediction model determining unit is specifically configured to determine the loss function based on the following formula:
；
wherein ,representing the loss value of the loss function, wherein alpha represents the overall loss function weight of the target at the moment of the first real track, and beta represents the overall loss function weight of the target at the moment of the second real track;respectively representing a first real track moment space target loss function, a first real track moment moving target loss function, a second real track moment space target loss function and a second real track moment moving target loss function; γ1, γ2, γ3, γ4 respectively represent the first real track moment spatial target loss function weight, the first real track moment moving target loss function weight, the second real track moment spatial target loss function weight, and the second real track moment moving target loss function weight.
Optionally, the first real track moment spatial target loss function includes a difference between the position information of the historical vehicle at the first historical predicted moment and the position information of the historical vehicle at the first real track moment; the first real track moment moving target loss function comprises a difference between the moving state information of the historical vehicle at a first historical predicting moment and the moving state information of the historical vehicle at the first real track moment; the second real track moment space target loss function comprises the difference between the position information of the historical vehicle at the second historical prediction moment and the position information of the historical vehicle at the second real track moment; the second real track moment moving target loss function comprises a difference between the moving state information of the historical vehicle at the second historical predicting moment and the moving state information of the historical vehicle at the second real track moment.
The vehicle path planning device provided by the embodiment of the invention can execute the vehicle path planning method provided by any embodiment of the invention, and has the corresponding functional modules and beneficial effects of the execution method.
Fig. 7 shows a schematic diagram of an electronic device 40 that may be used to implement an embodiment of the invention. Electronic devices are intended to represent various forms of digital computers, such as laptops, desktops, workstations, personal digital assistants, servers, blade servers, mainframes, and other appropriate computers. Electronic equipment may also represent various forms of mobile devices, such as personal digital processing, cellular telephones, smartphones, wearable devices (e.g., helmets, glasses, watches, etc.), and other similar computing devices. The components shown herein, their connections and relationships, and their functions, are meant to be exemplary only, and are not meant to limit implementations of the inventions described and/or claimed herein.
As shown in fig. 7, the electronic device 40 includes at least one processor 41, and a memory communicatively connected to the at least one processor 41, such as a Read Only Memory (ROM) 42, a Random Access Memory (RAM) 43, etc., in which the memory stores a computer program executable by the at least one processor, and the processor 41 may perform various suitable actions and processes according to the computer program stored in the Read Only Memory (ROM) 42 or the computer program loaded from the storage unit 48 into the Random Access Memory (RAM) 43. In the RAM 43, various programs and data required for the operation of the electronic device 40 may also be stored. The processor 41, the ROM 42 and the RAM 43 are connected to each other via a bus 44. An input/output (I/O) interface 45 is also connected to bus 44.
Various components in electronic device 40 are connected to I/O interface 45, including: an input unit 46 such as a keyboard, a mouse, etc.; an output unit 47 such as various types of displays, speakers, and the like; a storage unit 48 such as a magnetic disk, an optical disk, or the like; and a communication unit 49 such as a network card, modem, wireless communication transceiver, etc. The communication unit 49 allows the electronic device 40 to exchange information/data with other devices via a computer network, such as the internet, and/or various telecommunication networks.
The processor 41 may be various general and/or special purpose processing components with processing and computing capabilities. Some examples of processor 41 include, but are not limited to, a Central Processing Unit (CPU), a Graphics Processing Unit (GPU), various specialized Artificial Intelligence (AI) computing chips, various processors running machine learning model algorithms, digital Signal Processors (DSPs), and any suitable processor, controller, microcontroller, etc. The processor 41 performs the various methods and processes described above, such as a path planning method for a vehicle.
In some embodiments, the path planning method of the vehicle may be implemented as a computer program tangibly embodied on a computer-readable storage medium, such as the storage unit 48. In some embodiments, part or all of the computer program may be loaded and/or installed onto the electronic device 40 via the ROM 42 and/or the communication unit 49. When the computer program is loaded into the RAM 43 and executed by the processor 41, one or more steps of the path planning method of the vehicle described above may be performed. Alternatively, in other embodiments, the processor 41 may be configured to perform the path planning method of the vehicle in any other suitable way (e.g., by means of firmware).
Various implementations of the systems and techniques described here above may be implemented in digital electronic circuitry, integrated circuit systems, field Programmable Gate Arrays (FPGAs), application Specific Integrated Circuits (ASICs), application Specific Standard Products (ASSPs), systems On Chip (SOCs), load programmable logic devices (CPLDs), computer hardware, firmware, software, and/or combinations thereof. These various embodiments may include: implemented in one or more computer programs, the one or more computer programs may be executed and/or interpreted on a programmable system including at least one programmable processor, which may be a special purpose or general-purpose programmable processor, that may receive data and instructions from, and transmit data and instructions to, a storage system, at least one input device, and at least one output device.
A computer program for carrying out methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general purpose computer, special purpose computer, or other programmable data processing apparatus, such that the computer programs, when executed by the processor, cause the functions/acts specified in the flowchart and/or block diagram block or blocks to be implemented. The computer program may execute entirely on the machine, partly on the machine, as a stand-alone software package, partly on the machine and partly on a remote machine or entirely on the remote machine or server.
In the context of the present invention, a computer-readable storage medium may be a tangible medium that can contain, or store a computer program for use by or in connection with an instruction execution system, apparatus, or device. The computer readable storage medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. Alternatively, the computer readable storage medium may be a machine readable signal medium. More specific examples of a machine-readable storage medium would include an electrical connection based on one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.
To provide for interaction with a user, the systems and techniques described here can be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to a user; and a keyboard and a pointing device (e.g., a mouse or a trackball) through which a user can provide input to the electronic device. Other kinds of devices may also be used to provide for interaction with a user; for example, feedback provided to the user may be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form, including acoustic input, speech input, or tactile input.
The systems and techniques described here can be implemented in a computing system that includes a background component (e.g., as a data server), or that includes a middleware component (e.g., an application server), or that includes a front-end component (e.g., a user computer having a graphical user interface or a web browser through which a user can interact with an implementation of the systems and techniques described here), or any combination of such background, middleware, or front-end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: local Area Networks (LANs), wide Area Networks (WANs), blockchain networks, and the internet.
The computing system may include clients and servers. The client and server are typically remote from each other and typically interact through a communication network. The relationship of client and server arises by virtue of computer programs running on the respective computers and having a client-server relationship to each other. The server can be a cloud server, also called a cloud computing server or a cloud host, and is a host product in a cloud computing service system, so that the defects of high management difficulty and weak service expansibility in the traditional physical hosts and VPS service are overcome.
It should be appreciated that various forms of the flows shown above may be used to reorder, add, or delete steps. For example, the steps described in the present invention may be performed in parallel, sequentially, or in a different order, so long as the desired results of the technical solution of the present invention are achieved, and the present invention is not limited herein.
The above embodiments do not limit the scope of the present invention. It will be apparent to those skilled in the art that various modifications, combinations, sub-combinations and alternatives are possible, depending on design requirements and other factors. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the scope of the present invention.
Claims (10)
1. A method of path planning for a vehicle, comprising:
determining target data of a target vehicle at the current moment; the target data comprises vehicle body information, vehicle state information, barrier information and map information; the vehicle state information comprises position information and motion state information;
obtaining the own vehicle state information of the target vehicle at least one first prediction moment and the own vehicle state information of the target vehicle at a second prediction moment according to the target data and based on a prediction model; the first predicted time is earlier than the second predicted time; the prediction model is obtained by training based on historical data of at least one historical vehicle;
Determining a target running path of the target vehicle according to the target data, the self-vehicle state information at the first prediction time and the self-vehicle state information at the second prediction time;
wherein, the determining process of the prediction model comprises the following steps:
for each historical data, determining starting data of the historical vehicle according to the historical data; the initial data comprise vehicle body information, vehicle state information, barrier information and reference point information of the historical vehicle at the initial moment in the historical data;
inputting the initial data into a preset prediction algorithm to obtain the vehicle state information of the first historical prediction moment and the vehicle state information of the second historical prediction moment;
determining the vehicle state information of the historical vehicle at the first real track moment in the historical data;
determining the own vehicle state information of the historical vehicle at the second real track moment in the historical data;
and determining the prediction model according to the initial data, the own vehicle state information of each historical prediction moment historical vehicle, the own vehicle state information of each actual track moment historical vehicle and a loss function of a preset prediction algorithm.
2. The method according to claim 1, wherein the reference point is determined from the start data and a preset distance; the first real track moment is determined according to the starting moment and a preset period.
3. The method of claim 2, wherein the reference point determination comprises:
determining a target lane where the historical vehicle is located according to the historical data;
determining a target reference line of the target lane;
and determining a preset number of reference points from the target reference line according to the initial data and a preset distance.
4. The method of claim 1, wherein the determining of the second true track moment comprises:
taking the moment of a preset time period from the starting moment in the historical data as the moment of a second real track; or alternatively, the process may be performed,
and determining the orientation angle of each track point in the historical data in turn, determining the orientation angle difference value between the orientation angle of each track point and the orientation angle of the previous track point, and determining the moment of the track point in the historical data of the historical vehicle as the second real track moment when the absolute value of the orientation angle difference value is larger than a preset threshold value.
5. The method of claim 1, wherein the pre-set predictive algorithm comprises a self-attention algorithm; the network structure of the self-attention algorithm comprises a self-attention layer and a multi-head-multi-layer full-connection layer;
the determining process of the self-vehicle state information at the first historical predicting moment and the self-vehicle state information at the second historical predicting moment comprises the following steps:
vectorizing the vehicle body information and the vehicle state information in the initial data to obtain a first vector;
carrying out vectorization processing on the barrier information in the initial data to obtain a second vector;
vectorizing the reference point information in the initial data to obtain a third vector;
splicing the first vector, the second vector and the third vector to obtain a first target vector and a second target vector;
processing the first target vector, the first vector and the second target vector through the self-attention layer and the multi-head-multi-layer full-connection layer to obtain a target planning result of the historical vehicle at the first historical prediction moment and a target planning result of the historical vehicle at the second historical prediction moment; the target planning result comprises a space planning result and a motion planning result.
6. The method of claim 1, wherein the loss function is determined based on the following formula:
；
wherein , representing the loss value of the loss function, wherein alpha represents the overall loss function weight of the target at the moment of the first real track, and beta represents the overall loss function weight of the target at the moment of the second real track;respectively representing a first real track moment space target loss function, a first real track moment moving target loss function, a second real track moment space target loss function and a second real track moment moving target loss function; γ1, γ2, γ3, γ4 respectively represent the first real track moment spatial target loss function weight, the first real track moment moving target loss function weight, the second real track moment spatial target loss function weight, and the second real track moment moving target loss function weight.
7. The method of claim 6, wherein the first real track moment spatial target loss function includes a difference between the position information of the historical vehicle at a first historical predicted moment and the position information of the historical vehicle at a first real track moment; the first real track moment moving target loss function comprises a difference between the moving state information of the historical vehicle at a first historical predicting moment and the moving state information of the historical vehicle at the first real track moment; the second real track moment space target loss function comprises the difference between the position information of the historical vehicle at the second historical prediction moment and the position information of the historical vehicle at the second real track moment; the second real track moment moving target loss function comprises a difference between the moving state information of the historical vehicle at the second historical predicting moment and the moving state information of the historical vehicle at the second real track moment.
8. A path planning apparatus for a vehicle, comprising:
the target data determining module is used for determining target data of the target vehicle at the current moment; the target data comprises vehicle body information, vehicle state information, barrier information and map information; the vehicle state information comprises position information and motion state information;
the future time own vehicle state information determining module is used for obtaining own vehicle state information of the target vehicle at least one first prediction time and own vehicle state information of the target vehicle at a second prediction time according to the target data and based on a prediction model; the first predicted time is earlier than the second predicted time; the prediction model is obtained by training based on historical data of at least one historical vehicle;
the target driving path determining module is used for determining a target driving path of the target vehicle according to the target data, the vehicle state information at the first prediction moment and the vehicle state information at the second prediction moment;
the prediction model determining module comprises a starting data determining unit, a first prediction model determining unit and a second prediction model determining unit, wherein the starting data determining unit is used for determining, for each historical data, starting data of the historical vehicle according to the historical data before obtaining the self-vehicle state information of the target vehicle at least one first prediction moment and the self-vehicle state information of the target vehicle at a second prediction moment according to the target data and the prediction model; the initial data comprise vehicle body information, vehicle state information, barrier information and reference point information of the historical vehicle at the initial moment in the historical data;
The historical prediction moment self-vehicle state information determining unit is used for inputting the initial data into a preset prediction algorithm to obtain self-vehicle state information of the first historical prediction moment and self-vehicle state information of the second historical prediction moment;
a first vehicle state information determining unit of a historical vehicle is used for determining vehicle state information of the historical vehicle at a first real track moment in the historical data;
a second vehicle state information determining unit of the historical vehicle, configured to determine vehicle state information of the historical vehicle at a second real track moment in the historical data;
and the prediction model determining unit is used for determining the prediction model according to the initial data, the own vehicle state information of each historical prediction moment historical vehicle, the own vehicle state information of each actual track moment historical vehicle and a loss function of a preset prediction algorithm.
9. An electronic device, the electronic device comprising:
at least one processor; and
a memory communicatively coupled to the at least one processor; wherein,
the memory stores a computer program executable by the at least one processor to enable the at least one processor to perform the path planning method of the vehicle of any one of claims 1-7.
10. A computer readable storage medium storing computer instructions for causing a processor to perform the method of path planning for a vehicle according to any one of claims 1-7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310770396.7A CN116499487B (en) | 2023-06-28 | 2023-06-28 | Vehicle path planning method, device, equipment and medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310770396.7A CN116499487B (en) | 2023-06-28 | 2023-06-28 | Vehicle path planning method, device, equipment and medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116499487A CN116499487A (en) | 2023-07-28 |
| CN116499487B true CN116499487B (en) | 2023-09-05 |
Family
ID=87330610
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310770396.7A Active CN116499487B (en) | 2023-06-28 | 2023-06-28 | Vehicle path planning method, device, equipment and medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116499487B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116842392B (en) * | 2023-08-29 | 2024-04-16 | 新石器慧通（北京）科技有限公司 | Track prediction method and training method, device, equipment and medium of model thereof |
Citations (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109492835A (en) * | 2018-12-28 | 2019-03-19 | 东软睿驰汽车技术（沈阳）有限公司 | Determination method, model training method and the relevant apparatus of vehicle control information |
| CN109910880A (en) * | 2019-03-07 | 2019-06-21 | 百度在线网络技术（北京）有限公司 | Method, apparatus, storage medium and the terminal device of vehicle behavior planning |
| CN110908375A (en) * | 2019-11-14 | 2020-03-24 | 北京三快在线科技有限公司 | Method and device for acquiring lane change decision information, storage medium and vehicle |
| CN113128381A (en) * | 2021-04-06 | 2021-07-16 | 浙江吉利控股集团有限公司 | Obstacle trajectory prediction method, system and computer storage medium |
| CN113465609A (en) * | 2020-03-30 | 2021-10-01 | 阿里巴巴集团控股有限公司 | Time sequence matching method and device for target object |
| CN113561994A (en) * | 2021-08-13 | 2021-10-29 | 北京三快在线科技有限公司 | Trajectory planning method and device, storage medium and electronic equipment |
| CN113942524A (en) * | 2020-07-15 | 2022-01-18 | 广州汽车集团股份有限公司 | Vehicle running control method and system and computer readable storage medium |
| CN114323054A (en) * | 2022-01-12 | 2022-04-12 | 苏州挚途科技有限公司 | Method and device for determining running track of automatic driving vehicle and electronic equipment |
| CN114572244A (en) * | 2022-02-28 | 2022-06-03 | 上海和夏新能源科技有限公司 | Intelligent driving and ADAS front vehicle driving behavior recognition and prediction method and module |
| CN115447617A (en) * | 2022-11-10 | 2022-12-09 | 清华大学苏州汽车研究院（相城） | Vehicle control method, device, equipment and medium |
| WO2022257283A1 (en) * | 2021-06-08 | 2022-12-15 | 新石器慧通（北京）科技有限公司 | Vehicle obstacle avoidance method and apparatus, electronic device, and storage medium |
| CN116166030A (en) * | 2023-03-06 | 2023-05-26 | 中国工商银行股份有限公司 | Path planning method and device, storage medium and electronic equipment |
| CN116295496A (en) * | 2023-03-22 | 2023-06-23 | 新石器慧通（北京）科技有限公司 | Automatic driving vehicle path planning method, device, equipment and medium |
| CN116337093A (en) * | 2021-12-23 | 2023-06-27 | 北京嘀嘀无限科技发展有限公司 | Path planning method, device, equipment, storage medium and product |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US10908608B2 (en) * | 2018-01-18 | 2021-02-02 | Baidu Usa Llc | Method and system for stitching planning trajectories from consecutive planning cycles for smooth control execution of autonomous driving vehicles |
-
2023
- 2023-06-28 CN CN202310770396.7A patent/CN116499487B/en active Active
Patent Citations (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109492835A (en) * | 2018-12-28 | 2019-03-19 | 东软睿驰汽车技术（沈阳）有限公司 | Determination method, model training method and the relevant apparatus of vehicle control information |
| CN109910880A (en) * | 2019-03-07 | 2019-06-21 | 百度在线网络技术（北京）有限公司 | Method, apparatus, storage medium and the terminal device of vehicle behavior planning |
| CN110908375A (en) * | 2019-11-14 | 2020-03-24 | 北京三快在线科技有限公司 | Method and device for acquiring lane change decision information, storage medium and vehicle |
| CN113465609A (en) * | 2020-03-30 | 2021-10-01 | 阿里巴巴集团控股有限公司 | Time sequence matching method and device for target object |
| CN113942524A (en) * | 2020-07-15 | 2022-01-18 | 广州汽车集团股份有限公司 | Vehicle running control method and system and computer readable storage medium |
| CN113128381A (en) * | 2021-04-06 | 2021-07-16 | 浙江吉利控股集团有限公司 | Obstacle trajectory prediction method, system and computer storage medium |
| WO2022257283A1 (en) * | 2021-06-08 | 2022-12-15 | 新石器慧通（北京）科技有限公司 | Vehicle obstacle avoidance method and apparatus, electronic device, and storage medium |
| CN113561994A (en) * | 2021-08-13 | 2021-10-29 | 北京三快在线科技有限公司 | Trajectory planning method and device, storage medium and electronic equipment |
| CN116337093A (en) * | 2021-12-23 | 2023-06-27 | 北京嘀嘀无限科技发展有限公司 | Path planning method, device, equipment, storage medium and product |
| CN114323054A (en) * | 2022-01-12 | 2022-04-12 | 苏州挚途科技有限公司 | Method and device for determining running track of automatic driving vehicle and electronic equipment |
| CN114572244A (en) * | 2022-02-28 | 2022-06-03 | 上海和夏新能源科技有限公司 | Intelligent driving and ADAS front vehicle driving behavior recognition and prediction method and module |
| CN115447617A (en) * | 2022-11-10 | 2022-12-09 | 清华大学苏州汽车研究院（相城） | Vehicle control method, device, equipment and medium |
| CN116166030A (en) * | 2023-03-06 | 2023-05-26 | 中国工商银行股份有限公司 | Path planning method and device, storage medium and electronic equipment |
| CN116295496A (en) * | 2023-03-22 | 2023-06-23 | 新石器慧通（北京）科技有限公司 | Automatic driving vehicle path planning method, device, equipment and medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116499487A (en) | 2023-07-28 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113715814B (en) | Collision detection method, device, electronic equipment, medium and automatic driving vehicle | |
| CN112526999B (en) | Speed planning method, device, electronic equipment and storage medium | |
| CN113682318B (en) | Vehicle running control method and device | |
| CN116499487B (en) | Vehicle path planning method, device, equipment and medium | |
| CN114771572A (en) | Automatic driving track prediction method, device, equipment and storage medium | |
| CN114987546A (en) | Training method, device and equipment of trajectory prediction model and storage medium | |
| CN115675534A (en) | Vehicle track prediction method and device, electronic equipment and storage medium | |
| CN113753078B (en) | Obstacle track prediction method and device, electronic equipment and automatic driving vehicle | |
| CN117168488A (en) | Vehicle path planning method, device, equipment and medium | |
| CN116295496A (en) | Automatic driving vehicle path planning method, device, equipment and medium | |
| CN116358584A (en) | Automatic driving vehicle path planning method, device, equipment and medium | |
| CN115497322B (en) | Narrow road vehicle meeting method, device, equipment and storage medium | |
| CN115782876A (en) | Lane changing track generation method, device and equipment and automatic driving vehicle | |
| CN114919661B (en) | Parking control method, device, equipment and storage medium | |
| CN115583254A (en) | Path planning method, device and equipment and automatic driving vehicle | |
| CN114030483B (en) | Vehicle control method, device, electronic equipment and medium | |
| CN114590248B (en) | Method and device for determining driving strategy, electronic equipment and automatic driving vehicle | |
| CN114407916B (en) | Vehicle control and model training method and device, vehicle, equipment and storage medium | |
| CN117565894A (en) | Automatic driving vehicle control method, device, equipment and storage medium | |
| CN114030483A (en) | Vehicle control method, device, electronic apparatus, and medium | |
| CN116700065A (en) | Control method and device of unmanned equipment, electronic equipment and storage medium | |
| CN116534058A (en) | Vehicle running control method and device, vehicle and storage medium | |
| CN116295495A (en) | Automatic driving vehicle path planning method, device, equipment and medium | |
| CN117928577A (en) | Path planning method and device, electronic equipment and storage medium | |
| CN117647258A (en) | Path planning method, device, equipment and storage medium |
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 |