CN110986994A - Automatic lane change intention marking method based on high-noise vehicle track data - Google Patents
Automatic lane change intention marking method based on high-noise vehicle track data Download PDFInfo
- Publication number
- CN110986994A CN110986994A CN201911113106.1A CN201911113106A CN110986994A CN 110986994 A CN110986994 A CN 110986994A CN 201911113106 A CN201911113106 A CN 201911113106A CN 110986994 A CN110986994 A CN 110986994A
- Authority
- CN
- China
- Prior art keywords
- vehicle
- lane
- lane change
- frame
- noise
- 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.)
- Granted
Links
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
- G01C21/00—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00
- G01C21/26—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00 specially adapted for navigation in a road network
- G01C21/34—Route searching; Route guidance
- G01C21/36—Input/output arrangements for on-board computers
- G01C21/3626—Details of the output of route guidance instructions
- G01C21/3658—Lane guidance
Landscapes
- Engineering & Computer Science (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Automation & Control Theory (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Traffic Control Systems (AREA)
Abstract
The invention relates to a lane change intention automatic labeling method based on high-noise vehicle track data, which comprises the following steps of S1, collecting and preprocessing the tracking track data of surrounding vehicles; s2: finding a lane change point; s3: determining a lane change starting point; s4: verifying effective track changing tracks; s5: and outputting the marked lane change frame and the lane keeping frame. The invention uses the average change of the distance between the vehicle and the road center point in time as a mark for judging the start of lane change, not only considers the influence of noise, but also makes a robust definition on the nature of ever-changing lane change behavior.
Description
Technical Field
The invention belongs to the technical field of unmanned vehicles, and particularly relates to a lane change intention automatic labeling method based on high-noise vehicle track data.
Background
With the development of unmanned technology, unmanned vehicles need the ability to predict the future intent and trajectory of surrounding vehicles in order to safely and efficiently navigate complex traffic scenarios and make optimal decisions. The driving intentions of the surrounding vehicles not only have an important influence on the decision planning of the unmanned vehicle, but also influence each other in terms of driving intentions. Human drivers can make corresponding predictions of their driving intentions and future trajectories according to their own experiences and their current and past behaviors and interactions of surrounding vehicles, thereby making important driving decisions such as overtaking, decelerating or changing lanes. Existing driver assistance systems typically lack this predictive capability and give driving decisions to human drivers entirely. But such decision-making capability is essential for unmanned vehicles, and the intent and trajectory prediction of surrounding vehicles as important inputs also play an increasingly important role in present unmanned systems.
Existing intent prediction algorithms are generally classified into two types, rule-based algorithms and learning-based algorithms. The rule-based algorithm, most typically a "gap acceptance model," assumes that the driver's lane change motivation is based on the lead and lag gaps of the target lane, and the method assumes that the driver is inclined to make a lane change if the gaps reach a minimum acceptable value. Although the method has the characteristic of simply and conveniently judging the intention of the vehicle, the method needs a large amount of tedious and time-consuming parameter fine adjustment. The learning-based algorithm is used for learning a function or a network model for vehicle intention resolution from a large amount of data, a large amount of training data is usually required, the classification effect of the learning-based algorithm is generally better than that of the rule-based algorithm, and the learning-based algorithm is robust.
At present, how to obtain a large amount of labeled data becomes one of the difficulties of the method. Part of the existing learning-based algorithm uses a method of manually labeling public data sets (such as NGSIM US101), a great deal of labor is consumed, and the public data sets are generally collected by using a road fixed sensor and are greatly different from data collected on an unmanned vehicle in the aspects of accuracy and the like; some of the data generated by simulation is used as a training set, and has no practical application operability.
In a publication published in 2018 under the heading "left Vehicle Cooperative Lane-changing Behavior from underlying concepts in the NGSIM dataset" (author Su, Shuang & M ü ling, katharana & Dolan, John & planisamy, praven & Mudalige, Priyantha), an automatic labeling method was used to label the NGSIM dataset for Vehicle intent, as shown in fig. 6, with the change in the angle of the Vehicle heading to the road as a sign of the start of the change, but this is impractical in practice, which presents two problems 1. in practice vehicles are small at the time of change of Lane, resulting in a more difficult recognition of the change of angle under the noise being tracked, 2. this method requires fixing a threshold for judging the change of angle of each Vehicle, but the change of the angle of the position and the speed of each Vehicle is not very different, so it does not apply to the road in the case of changing Lane.
The present invention has been made in view of the above circumstances.
Disclosure of Invention
In order to solve the above problems in the prior art, an object of the present invention is to provide an automatic lane change intention labeling method based on high-noise vehicle trajectory data, which automatically labels a vehicle lane change behavior and a lane keeping behavior, so as to label vehicle tracking data with noise in a large scale, generate training data, and input the training data into a vehicle intention and trajectory prediction model based on learning. The method can identify and classify the lane change data in the vehicle tracking track data, so as to generate the training data taking the frame as the unit. The invention uses the average change of the distance between the vehicle and the road center point in time as a mark for judging the start of lane change, not only considers the influence of noise, but also makes a robust definition on the nature of ever-changing lane change behavior.
The technical scheme of the invention is as follows: a method for automatically marking lane change intention based on high-noise vehicle track data comprises the following steps,
s1, collecting and preprocessing the tracking track data of the surrounding vehicles;
s2: finding a lane change point;
s3: determining a lane change starting point;
s4: verifying effective track changing tracks;
s5: and outputting the marked lane change frame and the lane keeping frame.
Further, in step S1, the tracking trajectory data of the surrounding vehicle is generated by the unmanned sensing system at the time of driving test, and is generated by fusing sensing data such as a camera, a laser radar, and a millimeter wave radar, and then tracking the vehicle, including a map including road information, vehicle physical information (vehicle type, width, height), vehicle position (based on the Frenet road coordinate system, cartesian world coordinate system), lane where the vehicle is located, speed (based on the Frenet road coordinate system), and vehicle direction (based on the Frenet road coordinate system).
Further, in step S1, the preprocessing of the tracking trace data includes the following data filtering:
1) only selecting the track data with map information, and when the vehicle cannot be positioned to a known map, the vehicle track is not selected;
2) stationary vehicle trajectories are not selected;
3) only vehicle tracking trajectories within a certain distance from the unmanned vehicle are selected.
Further, in step S2, the time point when the vehicle crosses different lanes is found in units of vehicles, that is, the vehicle in frame n-1 is located as lane a, and the vehicle in frame n is located as lane B, and the frame n is the lane change point of the vehicle.
Further, the step S3 includes: a) one setting is made for the behavior of the lane-change vehicle: when the lane changing behavior of the vehicle starts, the vehicle can gradually move towards the target lane, namely the L transverse distance between the vehicle and the current lane is increased gradually, and the L transverse distance between the vehicle and the target lane is decreased gradually; b) average value L of the first five frames using L-direction coordinatesmeanAs a judgment basis, starting from the nth frame of the lane change point of the vehicle, calculating the nth-1 frame, the nth-2 frame, themeanAccording to the definition in a), the L-direction distance of the vehicle of the n-1 th frame relative to the current lane is smaller than the L-direction distance of the vehicle of the n-th frame relative to the current lane, the n-2 th frame is smaller than the n-3 th frame, and the like till the n-t frame, when the L ismeanNo longer decreasing or reaching the central line of the lane (L is 0), defining the n-t frame as the beginning of lane changeAnd (4) point.
Further, in step S4, the track-changing tracks that are too short or too long are discarded and not considered. And c) verifying whether the vehicle is still in the current lane or not for each frame in the step b), immediately ending the next lane change for recording t if the lane change occurs once again in the lane change and return push process, and discarding the track if the t is too short or too long.
The invention has the advantages that: the method has the advantages that the robustness is high when the lane changing track is extracted from the vehicle tracking track data with serious noise, a generalized definition is provided for the vehicle lane changing behavior, the effectiveness of marking is verified in the marking process, and a large number of correct and effective lane changing frames and lane keeping frames can be marked under the condition of no manual intervention.
Drawings
FIG. 1 is a flow chart of the automatic lane change intention labeling method of the present invention.
Fig. 2 is a schematic diagram of lane keeping tracks and lane changing tracks in the high-noise vehicle tracking data according to the present invention, wherein the left side is the lane keeping track and the right side is the lane changing track.
FIG. 3 is a schematic representation of a Cartesian coordinate system and a Frenet coordinate system.
FIG. 4 is a lane change frame determination for the lane change intent automatic labeling method of the present invention.
Fig. 5 is a lane keeping frame determination of the lane change intention automatic labeling method of the present invention.
Fig. 6 is a schematic diagram illustrating vehicle intention labeling of an NGSIM data set by an automatic labeling method in the prior art.
Detailed Description
The method for automatically labeling a lane change intention based on high-noise vehicle trajectory data according to the present invention will be further described with reference to fig. 1 to 5, and it should be noted that the embodiments described below with reference to the drawings are exemplary and are intended to be illustrative of the present invention and should not be construed as limiting the present invention.
Referring to fig. 1, a flow chart of an automatic lane-change intention labeling method based on high-noise vehicle track data according to the present invention is shown, the automatic lane-change intention labeling method comprises the following steps,
and S1, collecting and preprocessing the tracking track data of the surrounding vehicles. The tracking trajectory data of the surrounding vehicles is generated by the driveway test of the unmanned sensing system, and includes a map with road information, vehicle physical information (such as vehicle type, vehicle size such as length, width and height), vehicle position (based on a Frenet road coordinate system or a Cartesian world coordinate system), lane where the vehicle is located, speed (preferably based on the Frenet road coordinate system), vehicle direction (preferably based on the Frenet road coordinate system), and the like, and is generated by fusing sensing data such as a camera, a laser radar, a millimeter wave radar, and the like and then tracking the vehicle, and the tracking trajectory data of the surrounding vehicles is high-noise vehicle tracking trajectory data, and as shown in FIG. 2, tracks are kept and lane change trajectories for the lanes in the high-noise vehicle tracking data.
In the present invention, the Frenet road coordinate system is preferably used. A cartesian coordinate system is often used to describe the position of an object, but is not the best choice for unmanned driving. Unmanned vehicles typically make decisions to accelerate or change lanes based on the road. As an alternative solution to the cartesian coordinate system, the Frenet coordinate system uses the center line of the road as a reference line, and defines the coordinate system using the tangential direction and the normal direction of the reference line. As shown in fig. 3, the tangential distance S and the lateral distance L of the vehicle with respect to the center line of the road are coordinates of the vehicle on the Frenet coordinate system.
In S1, the following preprocessing is performed on the tracking trajectory data of the surrounding vehicle:
1) only selecting the track data with map information, and when the vehicle cannot be positioned to a known map, the vehicle track is not selected;
2) the vehicle lane change and vehicle road keeping behaviors are only meaningful for moving vehicles, so that a static vehicle track is not selected;
3) only vehicle tracking trajectories within a certain value, for example within 100 meters, from the unmanned vehicle are selected. Since the tracking data is too noisy beyond 100 meters to be suitable as training data.
S2: and finding a lane change point. And (3) taking the vehicle as a unit, searching time points when the vehicle crosses different lanes, namely positioning the vehicle at the (n-1) th frame as a lane A, positioning the vehicle at the nth frame as a lane B, and then setting the nth frame as a lane change point of the vehicle. A vehicle may have one or more lane change points along the entire trajectory, or the entire trajectory may be lane keeping and have no lane change points.
S3: and determining a lane change starting point. a) The invention makes basic settings for the behavior of lane-changing vehicles: when the lane changing behavior of the vehicle starts, the vehicle gradually moves towards the target lane, namely the transverse distance L of the vehicle from the current lane is increased gradually, and the transverse distance L from the target lane is decreased gradually. However, this is not necessarily true for noise-laden tracking data, and a large amount of "jumping" occurs in the positioning of the vehicle on the Frenet road coordinate system, not only in the S direction but also in the L direction. In fig. 4, the diamond shape should be the real track-changing starting point, but the corresponding positions of the rectangles would be mistaken for the track-changing starting point according to the setting, so that improvement is needed.
b) The invention provides a method for determining the moving trend of a vehicle, which uses the average value L of the first five frames of an L-direction coordinatemeanAs a judgment basis, the robustness of lane change behavior judgment is greatly increased: starting from the nth frame of the lane changing point of the vehicle (a left triangle in fig. 4), the nth-1 frame, the nth-2 frame, the equal L of the current lane are calculated forward in sequencemeanAccording to the definition in a), the L-direction distance of the vehicle of the n-1 th frame relative to the current lane is smaller than the L-direction distance of the vehicle of the n-th frame relative to the current lane, the n-2 th frame is smaller than the n-3 th frame, and the like till the n-t frame, when the L ismeanAnd when the vehicle does not decrease any more or reaches the center line of the lane (L is 0), defining the n-t frame as the starting point of the lane change of the vehicle. The curve on the more gradual right side in FIG. 4 is the one using the substitute LmeanThe new trace plot can be seen to be more robust to noise than the original trace next to the curve.
S4: and verifying the effective track changing track. Too short or too long lane change tracks (e.g., t <10 or t >100) are discarded from consideration; for each frame in b) of step S3, it is verified whether the vehicle is still in the current lane, if a lane change occurs again during the lane change pushback process, the pushback is ended immediately and t is recorded, and if t is too short or too long, the track is discarded. The verification can effectively filter out over-noisy or incomplete data. As shown in fig. 5, the left graph of fig. 5 is a lane change track, the triangle point at the front end is the actual lane change point, and the other triangles are the lane change points caused by noise. The right diagram of fig. 5 is a lane keeping track, and the triangle is a lane changing point caused by noise. By the step, both tracks can be discarded and not considered, thereby ensuring the correctness of the training data.
S5: and outputting the marked lane change frame and the lane keeping frame. All frames [ n-t, n ] from the start of a vehicle lane change to a lane change point are defined as vehicle lane change frames, and in addition, are lane keeping frames. The lane change end point is not set, and the lane change end point is the lane change point, because the actual behavior of the vehicle after the lane change point is lane keeping until the next lane change.
The technical scheme of the invention has strong robustness on extracting the lane change track from the vehicle tracking track data with serious noise, provides a generalized definition for the vehicle lane change behavior, verifies the effectiveness of the marking in the marking process, and can mark a large number of correct and effective lane change frames and lane keeping frames under the condition of no manual intervention.
The method of the present invention is not limited to autonomous vehicles and is equally applicable to trucks and other vehicle types.
The specific parameters of the invention can be adjusted, and the invention is also suitable for vehicle track data with low noise, and is not limited to tracking data.
The above-described embodiments are intended to illustrate rather than to limit the invention, and any modifications and variations of the present invention are within the spirit of the invention and the scope of the claims.
Claims (6)
1. A method for automatically marking lane change intention based on high-noise vehicle track data is characterized by comprising the following steps,
s1, collecting and preprocessing the tracking track data of the surrounding vehicles;
s2: finding a lane change point;
s3: determining a lane change starting point;
s4: verifying effective track changing tracks;
s5: and outputting the marked lane change frame and the lane keeping frame.
2. The method for automatically labeling a lane change intention based on high-noise vehicle trajectory data as claimed in claim 1, wherein in the step S1, the tracking trajectory data of the surrounding vehicles is generated by the unmanned sensing system during driving test, and comprises a map with road information, vehicle physical information, vehicle position, lane, speed and vehicle orientation.
3. The automatic marking method for lane change intention based on high-noise vehicle trajectory data as claimed in claim 1, wherein in the step S1, the preprocessing of the tracking trajectory data includes the following data screening:
1) only selecting the track data with map information, and when the vehicle cannot be positioned to a known map, the vehicle track is not selected;
2) stationary vehicle trajectories are not selected;
3) only vehicle tracking trajectories within a certain distance from the unmanned vehicle are selected.
4. The method as claimed in claim 1, wherein in step S2, the time points when the vehicle crosses different lanes are found in units of vehicles, that is, the n-1 th frame is located as lane a, the n-th frame is located as lane B, and the n-th frame is the lane change point of the vehicle.
5. The method for automatically labeling a lane change intention based on high-noise vehicle trajectory data according to claim 1, wherein the step S3 further comprises:
a) one setting is made for the behavior of the lane-change vehicle: when the lane changing behavior of the vehicle starts, the vehicle can gradually move towards the target lane, namely the L transverse distance between the vehicle and the current lane is increased gradually, and the L transverse distance between the vehicle and the target lane is decreased gradually;
b) average value L of the first five frames using L-direction coordinatesmeanAs a judgment basis, starting from the nth frame of the lane change point of the vehicle, calculating the nth-1 frame, the nth-2 frame, themeanAccording to the definition in a), the L-direction distance of the vehicle of the n-1 th frame relative to the current lane is smaller than the L-direction distance of the vehicle of the n-th frame relative to the current lane, the n-2 th frame is smaller than the n-3 th frame, and the like till the n-t frame, when the L ismeanAnd when the vehicle does not decrease any more or reaches the center line of the lane, defining the n-t frame as the starting point of the lane change of the vehicle.
6. The method for automatically labeling a lane change intention based on high-noise vehicle trajectory data according to claim 5, wherein in step S4, too short or too long lane change trajectories are discarded and not considered.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911113106.1A CN110986994B (en) | 2019-11-14 | 2019-11-14 | Automatic lane change intention marking method based on high-noise vehicle track data |
| PCT/CN2020/098164 WO2021093335A1 (en) | 2019-11-14 | 2020-06-24 | Method for automatically labeling lane changing intention based on high-noise trajectory data of vehicle |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911113106.1A CN110986994B (en) | 2019-11-14 | 2019-11-14 | Automatic lane change intention marking method based on high-noise vehicle track data |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110986994A true CN110986994A (en) | 2020-04-10 |
| CN110986994B CN110986994B (en) | 2021-08-03 |
Family
ID=70084361
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911113106.1A Active CN110986994B (en) | 2019-11-14 | 2019-11-14 | Automatic lane change intention marking method based on high-noise vehicle track data |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN110986994B (en) |
| WO (1) | WO2021093335A1 (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111599182A (en) * | 2020-07-22 | 2020-08-28 | 中汽院汽车技术有限公司 | Automatic lane changing behavior extraction method based on natural driving data |
| WO2021093335A1 (en) * | 2019-11-14 | 2021-05-20 | Suzhou Zhijia Science & Technologies Co., Ltd. | Method for automatically labeling lane changing intention based on high-noise trajectory data of vehicle |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115482659B (en) * | 2022-08-18 | 2023-09-19 | 浙江工商大学 | Intelligent autonomous decision-making method based on deep reinforcement learning |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103823382A (en) * | 2014-02-27 | 2014-05-28 | 浙江省科威工程咨询有限公司 | Lane change track optimization and visualization achievement method based on vehicle models and vehicle speeds |
| CN105667516A (en) * | 2014-12-03 | 2016-06-15 | 现代摩比斯株式会社 | System and method for estimating driving lane of vehicle |
| CN107415945A (en) * | 2016-05-24 | 2017-12-01 | 通用汽车环球科技运作有限责任公司 | For assessing the automotive drive system and its application method of track lane-change |
| CN107545612A (en) * | 2016-06-23 | 2018-01-05 | 戚艳红 | Make video recording startup method and the activation system of making video recording of a kind of drive recorder |
| CN108693883A (en) * | 2018-06-06 | 2018-10-23 | 西安智加科技有限公司 | A kind of the high-precision parking method and system of Vehicular automatic driving |
| JP2019059451A (en) * | 2017-09-28 | 2019-04-18 | トヨタ自動車株式会社 | Drive assist apparatus |
| CN109946688A (en) * | 2019-03-18 | 2019-06-28 | 中国汽车工程研究院股份有限公司 | Lane-change contextual data extracting method, device and server |
| CN110103955A (en) * | 2019-04-28 | 2019-08-09 | 浙江吉利控股集团有限公司 | A kind of vehicle early warning method, device and electronic equipment |
| CN110304075A (en) * | 2019-07-04 | 2019-10-08 | 清华大学 | Track of vehicle prediction technique based on Mix-state DBN and Gaussian process |
| CN110329263A (en) * | 2019-07-05 | 2019-10-15 | 西南交通大学 | The adaptive lane-change method for planning track of automatic driving vehicle |
Family Cites Families (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8972077B2 (en) * | 2012-06-20 | 2015-03-03 | Trimble Navigation Limited | Lane change monitoring |
| JP6815958B2 (en) * | 2017-09-19 | 2021-01-20 | トヨタ自動車株式会社 | Vehicle peripheral monitoring device |
| CN107967486B (en) * | 2017-11-17 | 2020-08-28 | 江苏大学 | Method for recognizing behaviors of surrounding vehicles |
| CN110146100B (en) * | 2018-02-13 | 2021-08-13 | 华为技术有限公司 | Trajectory prediction method, apparatus and storage medium |
| CN109886304B (en) * | 2019-01-22 | 2023-09-29 | 江苏大学 | HMM-SVM double-layer improved model-based surrounding vehicle behavior recognition method under complex road conditions |
| CN110986994B (en) * | 2019-11-14 | 2021-08-03 | 苏州智加科技有限公司 | Automatic lane change intention marking method based on high-noise vehicle track data |
-
2019
- 2019-11-14 CN CN201911113106.1A patent/CN110986994B/en active Active
-
2020
- 2020-06-24 WO PCT/CN2020/098164 patent/WO2021093335A1/en active Application Filing
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103823382A (en) * | 2014-02-27 | 2014-05-28 | 浙江省科威工程咨询有限公司 | Lane change track optimization and visualization achievement method based on vehicle models and vehicle speeds |
| CN105667516A (en) * | 2014-12-03 | 2016-06-15 | 现代摩比斯株式会社 | System and method for estimating driving lane of vehicle |
| CN107415945A (en) * | 2016-05-24 | 2017-12-01 | 通用汽车环球科技运作有限责任公司 | For assessing the automotive drive system and its application method of track lane-change |
| CN107545612A (en) * | 2016-06-23 | 2018-01-05 | 戚艳红 | Make video recording startup method and the activation system of making video recording of a kind of drive recorder |
| JP2019059451A (en) * | 2017-09-28 | 2019-04-18 | トヨタ自動車株式会社 | Drive assist apparatus |
| CN108693883A (en) * | 2018-06-06 | 2018-10-23 | 西安智加科技有限公司 | A kind of the high-precision parking method and system of Vehicular automatic driving |
| CN109946688A (en) * | 2019-03-18 | 2019-06-28 | 中国汽车工程研究院股份有限公司 | Lane-change contextual data extracting method, device and server |
| CN110103955A (en) * | 2019-04-28 | 2019-08-09 | 浙江吉利控股集团有限公司 | A kind of vehicle early warning method, device and electronic equipment |
| CN110304075A (en) * | 2019-07-04 | 2019-10-08 | 清华大学 | Track of vehicle prediction technique based on Mix-state DBN and Gaussian process |
| CN110329263A (en) * | 2019-07-05 | 2019-10-15 | 西南交通大学 | The adaptive lane-change method for planning track of automatic driving vehicle |
Non-Patent Citations (2)
| Title |
|---|
| SHUANG SU 等: "Learning Vehicle Cooperative Lane-changing Behavior from Observed Trajectories in the NGSIM Dataset", 《THE 29TH IEEE INTELLIGENT VEHICLES SYMPOSIUM (IV 2018)》 * |
| 王建群 等: "辅助驾驶中的换道决策安全研究", 《安全与环境学报》 * |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2021093335A1 (en) * | 2019-11-14 | 2021-05-20 | Suzhou Zhijia Science & Technologies Co., Ltd. | Method for automatically labeling lane changing intention based on high-noise trajectory data of vehicle |
| CN111599182A (en) * | 2020-07-22 | 2020-08-28 | 中汽院汽车技术有限公司 | Automatic lane changing behavior extraction method based on natural driving data |
| CN111599182B (en) * | 2020-07-22 | 2020-10-20 | 中汽院汽车技术有限公司 | Automatic lane changing behavior extraction method based on natural driving data |
Also Published As
| Publication number | Publication date |
|---|---|
| WO2021093335A1 (en) | 2021-05-20 |
| CN110986994B (en) | 2021-08-03 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| JP7106664B2 (en) | Intelligent driving control method and device, electronic device, program and medium | |
| CN109657355B (en) | Simulation method and system for vehicle road virtual scene | |
| CN110843789B (en) | Vehicle lane change intention prediction method based on time sequence convolution network | |
| Berriel et al. | Ego-lane analysis system (elas): Dataset and algorithms | |
| US9053433B2 (en) | Assisting vehicle guidance over terrain | |
| CN110986994B (en) | Automatic lane change intention marking method based on high-noise vehicle track data | |
| CN113127466B (en) | Vehicle track data preprocessing method and computer storage medium | |
| EP2899669A1 (en) | Lane relative position estimation method and system for driver assistance systems | |
| US11420623B2 (en) | Systems for determining object importance in on-road driving scenarios and methods thereof | |
| CN114454878B (en) | Method and device for determining vehicle speed control model training sample | |
| CN115618932A (en) | Traffic incident prediction method and device based on internet automatic driving and electronic equipment | |
| KR102592830B1 (en) | Apparatus and method for predicting sensor fusion target in vehicle and vehicle including the same | |
| EP2405383A1 (en) | Assisting with guiding a vehicle over terrain | |
| CN114371015B (en) | Automatic driving test method, automatic driving test device, computer equipment and storage medium | |
| CN115292435A (en) | High-precision map updating method and device, electronic equipment and storage medium | |
| US20230281424A1 (en) | Method for Extracting Features from Data of Traffic Scenario Based on Graph Neural Network | |
| CN111066024B (en) | Method and device for recognizing a lane, driver assistance system and vehicle | |
| CN115027497B (en) | Target vehicle cut-in intention prediction method and readable storage medium | |
| CN114693722B (en) | Vehicle driving behavior detection method, detection device and detection equipment | |
| CN117058634A (en) | Expressway scene self-adaptive traffic offence behavior identification method | |
| CN114461671A (en) | Data processing method, vehicle, and computer-readable storage medium | |
| CN114842660A (en) | Unmanned lane track prediction method and device and electronic equipment | |
| US20240020964A1 (en) | Method and device for improving object recognition rate of self-driving car | |
| Sonata et al. | Street View Object Detection for Autonomous Car Steering Angle Prediction Using Convolutional Neural Network | |
| Siebinga | Communication-Enabled Interactions in Highway Traffic: A joint driver model for merging |
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 |