CN114550442A - Automatic driving vehicle state evaluation and early warning method based on vehicle-road cooperation - Google Patents
Automatic driving vehicle state evaluation and early warning method based on vehicle-road cooperation Download PDFInfo
- Publication number
- CN114550442A CN114550442A CN202111675077.5A CN202111675077A CN114550442A CN 114550442 A CN114550442 A CN 114550442A CN 202111675077 A CN202111675077 A CN 202111675077A CN 114550442 A CN114550442 A CN 114550442A
- Authority
- CN
- China
- Prior art keywords
- vehicle
- data
- automatic driving
- road
- equal
- 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
- 238000000034 method Methods 0.000 title claims abstract description 51
- 238000011156 evaluation Methods 0.000 title claims abstract description 43
- 230000003238 somatosensory effect Effects 0.000 claims abstract description 36
- 206010000117 Abnormal behaviour Diseases 0.000 claims abstract description 12
- 238000012545 processing Methods 0.000 claims abstract description 11
- 230000008569 process Effects 0.000 claims abstract description 8
- 238000012549 training Methods 0.000 claims abstract description 5
- 230000001133 acceleration Effects 0.000 claims description 58
- 238000001914 filtration Methods 0.000 claims description 18
- 230000002159 abnormal effect Effects 0.000 claims description 14
- 230000036461 convulsion Effects 0.000 claims description 14
- 238000004364 calculation method Methods 0.000 claims description 13
- 230000008859 change Effects 0.000 claims description 12
- 238000001514 detection method Methods 0.000 claims description 9
- 238000009826 distribution Methods 0.000 claims description 9
- 230000006870 function Effects 0.000 claims description 9
- 238000005516 engineering process Methods 0.000 claims description 7
- 238000005352 clarification Methods 0.000 claims description 3
- 238000003062 neural network model Methods 0.000 claims description 3
- 238000012216 screening Methods 0.000 claims description 3
- 238000007619 statistical method Methods 0.000 claims description 3
- 230000008447 perception Effects 0.000 claims 2
- 238000000605 extraction Methods 0.000 abstract description 3
- 238000011161 development Methods 0.000 description 2
- 238000013528 artificial neural network Methods 0.000 description 1
- 230000006399 behavior Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004422 calculation algorithm Methods 0.000 description 1
- 230000000295 complement effect Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
- 238000000844 transformation Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/01—Detecting movement of traffic to be counted or controlled
- G08G1/0104—Measuring and analyzing of parameters relative to traffic conditions
- G08G1/0125—Traffic data processing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/01—Detecting movement of traffic to be counted or controlled
- G08G1/0104—Measuring and analyzing of parameters relative to traffic conditions
- G08G1/0108—Measuring and analyzing of parameters relative to traffic conditions based on the source of data
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/01—Detecting movement of traffic to be counted or controlled
- G08G1/0104—Measuring and analyzing of parameters relative to traffic conditions
- G08G1/0108—Measuring and analyzing of parameters relative to traffic conditions based on the source of data
- G08G1/0116—Measuring and analyzing of parameters relative to traffic conditions based on the source of data from roadside infrastructure, e.g. beacons
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/01—Detecting movement of traffic to be counted or controlled
- G08G1/04—Detecting movement of traffic to be counted or controlled using optical or ultrasonic detectors
-
- 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
- Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Analytical Chemistry (AREA)
- Chemical & Material Sciences (AREA)
- Computational Linguistics (AREA)
- General Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Health & Medical Sciences (AREA)
- Traffic Control Systems (AREA)
Abstract
The invention discloses an automatic driving vehicle state evaluation and early warning method based on vehicle-road cooperation, which relates to the technical field of road traffic, and comprises the following two parts: evaluating the state of the occupant somatosensory comfort level of the automatic driving vehicle in the whole driving process, evaluating the safety state of the occupant and the vehicle in the whole driving process of the automatic driving vehicle, acquiring the driving data of the automatic driving vehicle and the non-automatic driving vehicle through a road end, fusing the driving data acquired by the automatic driving vehicle end, performing data processing and feature extraction, establishing an objective evaluation standard of the occupant somatosensory comfort level of the automatic driving vehicle, and using the objective evaluation standard as the state constraint of path planning; through the learning and training of the mass data, the abnormal behavior of the automatic driving vehicle can be predicted on line, when the unsafe state of the vehicle is predicted, the vehicle safety personnel or a remote monitoring center can be warned, the vehicle can take over in time, and the vehicle is separated from the unsafe state.
Description
Technical Field
The invention relates to the technical field of road traffic, in particular to an automatic driving vehicle state evaluation and early warning method based on vehicle-road cooperation.
Background
The autonomous vehicle state evaluation includes: 1. evaluating the state of the occupant somatosensory comfort level in the whole driving process of the automatic driving vehicle; 2. and evaluating the safety states of passengers and the vehicle in the whole automatic driving process. At present, only a vehicle end is used for collecting driving data, compared with data collected by a road end, the collected data is single and not universal, and the data quantity is small, so that the data characteristic extraction is not facilitated; comfort evaluation depends on the established vehicle comfort prediction model, prediction accuracy cannot be guaranteed, and the comfort evaluation is greatly influenced by factors such as road surface quality and gradient; the comfort evaluation standard is subjective evaluation, and an objective and referable evaluation system is lacked; emergency safety precautions can only occur when the vehicle is in an abnormal state, and such takeover generally lacks environmental suitability.
Disclosure of Invention
Aiming at the technical problems, the invention overcomes the defects of the prior art and provides an automatic driving vehicle state evaluation and early warning method based on vehicle-road cooperation, which comprises the following steps:
s1, data acquisition based on vehicle and road cooperation technology
S1.1, for a road side, acquiring driving data of an automatic driving vehicle and a non-automatic driving vehicle passing through the road section through a sensing system of a road side unit, and uploading the driving data to a cloud end platform through a road side unit network layer;
s1.2, for the side of the vehicle, the automatic driving vehicle is driven by a driver, vehicle driving data are collected by means of a vehicle-mounted sensing layer, and the driving data are transmitted to a cloud platform through a vehicle-mounted network terminal;
s2, somatosensory comfort state assessment
S2.1 somatosensory comfort evaluation index: accx: longitudinal acceleration, m/s2(ii) a decx: longitudinal deceleration, m/s2(ii) a accy: lateral acceleration, m/s2(ii) a jerkx: longitudinal jerk, m/s3(ii) a jerky: transverse jerk, m/s3；
S2.2, calculating the somatosensory comfort data;
s2.3, obtaining comfort index standard reference values under different speeds and curvatures according to different rules;
s2.4, on-line evaluation of somatosensory comfort: designing a hysteresis logic and a hysteresis curved surface, comparing the somatosensory data value with a standard somatosensory data hysteresis interval value in an online mode, and outputting the specific type of the uncomfortable state, the frequency of the uncomfortable state in the process of one-time automatic driving and the proportion of the uncomfortable time to the total driving time;
s3, safety state evaluation and early warning
A large amount of data covered by multiple scenes acquired by a vehicle-road cooperation technology is used as a data set for offline training of a neural network model, the model is output to judge whether abnormal behaviors occur in a vehicle within a period of time in the future, and the model is used for online prediction of self-driven vehicles and social vehicles within a certain range.
The technical scheme of the invention is further defined as follows:
in the aforementioned method for evaluating and warning the state of an automatically driven vehicle based on vehicle-road cooperation, in S1.1, the driving data includes a plurality of items of information including a position of a coordinate point of the vehicle, a longitudinal speed and an acceleration, a lateral speed and an acceleration, a vehicle course angle, a road curvature, and a steering radius in a global coordinate system.
In the method for evaluating and early warning the state of the automatic driving vehicle based on the vehicle-road cooperation, in S1.1, the sensing system comprises the laser radar, the camera and the millimeter wave radar, and in S1.2, the vehicle-mounted sensing layer comprises the IMU, the laser radar and the millimeter wave radar.
In the automatic driving vehicle state evaluation and early warning method based on vehicle-road cooperation, S2.2 somatosensory comfort degree data calculation:
accx: vehicle end: calculating according to the position coordinate information difference to obtain longitudinal acceleration, performing second-order low-pass filtering to obtain est _ accx, performing second-order low-pass filtering to the acceleration information obtained based on the IMU to obtain IMU _ accx, adding a confidence ratio to the two acceleration values to obtain ax _ filtered Ka est _ accx + Kb IMU _ accx, and performing zero-crossing detection to obtain a final longitudinal acceleration value accx; road end: according to the position coordinate information, carrying out differential calculation to obtain longitudinal acceleration, carrying out second-order low-pass filtering to obtain est _ accx, and carrying out zero-crossing detection to obtain a final longitudinal acceleration value accx;
decx: the method is consistent with the accx calculation method, and a final longitudinal acceleration value decx is obtained through zero-crossing detection;
accy: calculating to obtain curvature information kappa of the current running road as dyaw/ds according to the position information, wherein dyaw is the change of a course angle in unit time, and ds is the change of a vehicle position in unit time; vehicle end: performing first-order low-pass filtering on the lateral acceleration data based on the IMU to obtain IMU _ accy; road end: calculating est _ acy by using a position difference method;
jerkx: carrying out difference based on the longitudinal acceleration to obtain a longitudinal acceleration degree change rate jerk, and carrying out mean value filtering to obtain est _ jerkx;
jerky: and carrying out difference based on the lateral acceleration to obtain lateral acceleration change rate jerk, and carrying out robust mean filtering to obtain est _ jerky.
In the above automatic driving vehicle state assessment and early warning method based on vehicle-road cooperation, S2.3 somatosensory comfort evaluation criteria:
(1) processing vehicle driving data obtained by the vehicle road cooperation, calculating the curvature of a vehicle driving route and the current vehicle speed at each moment, and obtaining various driving comfort index values under the curvature-vehicle speed, so that a huge data lattice set is formed, and five kinds of curvature-speed-comfort index scattered point distributions can be obtained by separating and processing the data lattice set;
(2) the speed anchor point and the curvature anchor point are defined, the minimum turning diameter of the passenger car is 9.0-12.0m, and the curvature interval is designed to be 0-0.2m-1The interval size is 0.02, namely [0:0.02:0.2 ]]The designed vehicle speed interval is 0-22m/s, and the interval size is 1.0, namely [0:1:22 ]]；
(3) Screening out bad data, traversing speed anchor points and curvature anchor points, calculating extreme values, average values, mean square deviations, 25%, 50% and 75% data value distribution points of various comfort indexes in the range of anchor points, and fitting the data by using a mobile least square method;
(4) comparing and analyzing the vehicle end data, the road end data and the vehicle road cooperative data;
(5) and obtaining comfort index standard reference values under different speeds and curvatures according to different rules.
S2.3(5), designing an evaluation function, or modifying a penalty coefficient to obtain a proper reference standard value, or taking a Gaussian function as a membership function and then carrying out clarification processing to obtain the reference standard value.
In the above automatic driving vehicle state assessment and early warning method based on vehicle-road cooperation, S2.3, a 75% value distribution point is selected as a comfort level reference standard value.
S2.4, carrying out interpolation calculation by adopting a curved surface interpolation method to obtain the hysteresis upper limit and the hysteresis lower limit of the standard somatosensory data under the current kappa and speed, and carrying out online evaluation on the somatosensory comfort level according to the obtained standard interval and logic.
In the aforementioned method for evaluating and warning the state of the automatically driven vehicle based on the vehicle-road cooperation, S2 obtains the total comfort evaluation comparison grade corresponding to the discomfort times and the discomfort time ratio according to the fuzzy statistical method, specifically as follows:
(1) the automatic driving time is less than or equal to 10mins
The discomfort times are less than or equal to 6 times, the discomfort time proportion is less than or equal to 10 percent, and the comfort level is high;
the discomfort times are less than or equal to 18 times, the discomfort time proportion is less than or equal to 25 percent, and the comfort level is medium;
the discomfort times are less than or equal to 30 times, the discomfort time proportion is less than or equal to 40 percent, and the comfort level is low;
(2) the automatic driving time is less than or equal to 20mins
The discomfort times are less than or equal to 15 times, the discomfort time proportion is less than or equal to 15 percent, and the comfort level is high;
the discomfort times are less than or equal to 45 times, the discomfort time proportion is less than or equal to 30 percent, and the comfort level is medium;
the discomfort times are less than or equal to 80, the discomfort time proportion is less than or equal to 45 percent, and the comfort level is low;
(3) the automatic driving time is less than or equal to 40mins
The discomfort times are less than or equal to 45 times, the proportion of the discomfort time is less than or equal to 20 percent, and the comfort level is high;
the discomfort times are less than or equal to 150 times, the discomfort time proportion is less than or equal to 35 percent, and the comfort level is medium;
the uncomfortable times are less than or equal to 300 times, the uncomfortable time proportion is less than or equal to 50 percent, and the comfort level is low.
The automatic driving vehicle state assessment and early warning method based on vehicle-road cooperation comprises the third step of judging abnormal behaviors including abnormal parking, abnormal acceleration, abnormal deceleration, large-amplitude rotation of an abnormal steering wheel and collision risks.
The invention has the beneficial effects that:
(1) according to the method, the driving data of the automatic driving vehicle and the non-automatic driving vehicle are collected by the road end, the driving data collected by the automatic driving vehicle end are fused, data processing and characteristic extraction are carried out, and an objective evaluation standard for the somatosensory comfort of passengers of the automatic driving vehicle is established and can be used as the state constraint of path planning; through the learning and training of the mass data, the abnormal behavior of the automatic driving vehicle can be predicted on line, when the unsafe state of the vehicle is predicted, the vehicle safety personnel or a remote monitoring center is warned, the vehicle is taken over in time, and the vehicle is separated from the unsafe state;
(2) the invention fully utilizes the advantages of the quantity, the stability and the full-time work of the road end equipment, collects the vehicle running information under a specific road section through the road end side, can collect the driving data of a plurality of social vehicles, and has the advantages of wide sampling coverage, large data volume and high efficiency; the data are acquired through the vehicle end side, so that the method has the advantages of complex driving scene and accurate data acquisition; therefore, the data can be supplemented mutually by a data acquisition method of the vehicle-road cooperation technology, a large amount of driving data can be rapidly acquired, and the data can be contrasted and analyzed;
(3) the invention depends on massive vehicle running data, fully analyzes and statistically processes the data, establishes an objective evaluation system of longitudinal acceleration, longitudinal deceleration, transverse acceleration, longitudinal jerk and transverse jerk of the automatic driving vehicle under different speeds and different curvatures, and digitalizes comfort evaluation;
(4) the invention learns and trains data to obtain a vehicle behavior prediction model, predicts and outputs the dynamic state of the vehicle, and when abnormal behaviors such as abnormal parking, sudden acceleration, violent steering wheel hitting and the like deviating from a planned track occur in the prediction of the vehicle or the collision risk of surrounding vehicles on the vehicle is predicted, early warning information is early warned to a vehicle safety personnel or an uploading remote monitoring center, so that the vehicle and passengers are ensured to be safe in a real-time panoramic intervention manner.
Drawings
FIG. 1 is a data acquisition method based on a vehicle-road cooperation technique;
FIG. 2 is a method for establishing somatosensory comfort evaluation criteria;
FIG. 3 is a comfort evaluation criterion;
FIG. 4 is a method for on-line assessment of somatosensory comfort;
FIG. 5 shows the result of decx interpolation hysteresis interval.
Detailed Description
The automatic driving vehicle state assessment and early warning method based on vehicle-road cooperation provided by the embodiment specifically comprises the following steps:
data acquisition based on vehicle-road cooperation technology
As shown in figure 1 of the drawings, in which,
for the road side, the driving data (including a plurality of items of information such as a vehicle coordinate point position, a longitudinal speed and an acceleration, a transverse speed and an acceleration, a vehicle course angle, a road curvature, a turning radius and the like) of an automatic driving vehicle and a non-automatic driving vehicle passing through the road section are collected through a sensing system (such as a laser radar, a camera and a millimeter wave radar) of a road side unit of the road section, and the driving data are uploaded to a cloud end platform through a network layer of the road side unit.
And (II) for the side of the vehicle side, the automatic driving vehicle is driven by a driver, vehicle driving data are collected by means of a vehicle-mounted sensing layer (IMU, a laser radar, a millimeter wave radar and the like), and the driving data are transmitted to a cloud platform through a vehicle-mounted network terminal.
Second, somatosensory comfort state assessment
For an automatic driving vehicle, the aim of realizing automatic driving of the vehicle in various scenes is pursued, the index for evaluating the quality of the automatic driving comprises the comfort feeling of passengers, the driving comfort has different evaluation standards for different people and is subjective personal feeling.
Somatosensory comfort evaluation index
accx: longitudinal acceleration, m/s2；
decx: longitudinal deceleration, m/s2；
accy: lateral acceleration, m/s2；
jerkx: longitudinal jerk, m/s3；
jerky: transverse jerk, m/s3；
In the transverse direction, the left and right body feeling of the passengers is not poor due to the left and right symmetry, so that the left and right are not distinguished; in the longitudinal direction, the body feeling of the member is greatly different between the forward direction and the backward direction, so that the longitudinal acceleration is subdivided into the forward acceleration and the backward acceleration.
(II) somatosensory comfort data calculation
accx: vehicle end: calculating according to the position coordinate information difference to obtain longitudinal acceleration, performing second-order low-pass filtering to obtain est _ accx, performing second-order low-pass filtering to the acceleration information obtained based on the IMU to obtain IMU _ accx, adding a confidence ratio to the two acceleration values to obtain ax _ filtered Ka est _ accx + Kb IMU _ accx, and performing zero-crossing detection to obtain a final longitudinal acceleration value accx; road end: according to the position coordinate information, carrying out differential calculation to obtain longitudinal acceleration, carrying out second-order low-pass filtering to obtain est _ accx, and carrying out zero-crossing detection to obtain a final longitudinal acceleration value accx;
decx: the method is consistent with an accx calculation method, and a final longitudinal acceleration value decx is obtained through zero-crossing detection;
accy: calculating to obtain curvature information kappa of the current running road as dyaw/ds according to the position information, wherein dyaw is the change of a course angle in unit time, and ds is the change of a vehicle position in unit time; vehicle end: performing first-order low-pass filtering on the lateral acceleration data based on the IMU to obtain IMU _ accy; road end: calculating est _ acy by using a position difference method;
jerkx: carrying out difference based on the longitudinal acceleration to obtain a longitudinal acceleration degree change rate jerk, and carrying out mean value filtering to obtain est _ jerkx;
jerky: and carrying out difference based on the lateral acceleration to obtain lateral acceleration change rate jerk, and carrying out robust mean filtering to obtain est _ jerky.
(III) evaluation standard for somatosensory comfort level
As shown in figure 2 of the drawings, in which,
(1) processing vehicle driving data obtained by the vehicle road cooperation, calculating the curvature of a vehicle driving route and the current vehicle speed at each moment, and obtaining various driving comfort index values under the curvature-vehicle speed, so that a huge data lattice set is formed, and five kinds of curvature-speed-comfort index scattered point distributions can be obtained by separating and processing the data lattice set;
(2) the speed anchor point and the curvature anchor point are defined, the minimum turning diameter of the passenger car is 9.0-12.0m, and the curvature interval is designed to be 0-0.2m-1Interval size 0.02, [0:0.02:0.2]The designed vehicle speed interval is 0-22m/s, and the interval size is 1.0, namely [0:1:22 ]]；
(3) Screening bad data, traversing speed anchor points and curvature anchor points, calculating extreme values, average values, mean square deviations, 25%, 50% and 75% data value distribution points of various comfort indexes in an anchor point interval range, and fitting the data by using a mobile least square method;
(4) comparing and analyzing the vehicle end data, the road end data and the vehicle road cooperative data;
(5) the comfort index standard reference values under different speeds and curvatures are obtained according to different rules, for example, a proper reference standard value can be obtained by designing an evaluation function and modifying a penalty coefficient, and the reference standard value can also be obtained by performing clarification processing after a Gaussian function is used as a membership function.
The invention selects 75% of the numerical distribution points as comfort reference standard values, and obtains the result as shown in figure 3.
(IV) somatosensory comfort online assessment
With the objective evaluation standard of driving comfort, whether the driving is in a comfortable driving state can be judged according to the standard, but when the driving comfort index obtained by calculation fluctuates around the standard value, the judgment result is vibrated, and the judgment cannot be accurately carried out, so that hysteresis logic and a hysteresis curved surface are designed. As shown in fig. 4, in the online mode, current kappa and speed and corresponding comfort index values are calculated according to position, speed and acceleration information, these somatosensory data values are compared with the standard somatosensory data hysteresis interval values, the standard somatosensory data hysteresis upper limit and the standard somatosensory data hysteresis lower limit under the current kappa and speed are obtained by interpolation calculation with a curved surface interpolation method, and as shown in fig. 5, the hysteresis interval obtained by interpolation is obtained when comfort is evaluated in real time with decx as an example. According to the obtained standard interval and logic, the somatosensory comfort level can be evaluated on line to output the specific type of the uncomfortable state, the frequency of the uncomfortable state in the process of one-time automatic driving and the proportion of the uncomfortable time to the total driving time.
Further, an overall comfort evaluation comparison grade table corresponding to the discomfort times and the discomfort time proportion can be obtained according to a fuzzy statistical method, for example, as follows:
thirdly, safety state assessment and early warning
A large amount of data covered by multiple scenes acquired by a vehicle-road cooperation technology is used as a data set for offline training of a neural network model, the model is output to determine whether abnormal behaviors occur in a vehicle within a period of time in the future, and the abnormal behaviors comprise abnormal parking, abnormal acceleration, abnormal deceleration, large-amplitude rotation of an abnormal steering wheel and collision risks; the model is used for online prediction of self-automatic driving vehicles and social vehicles within a certain range, if abnormal behaviors of the self-vehicle or collision threats of surrounding vehicles to the self-vehicle are predicted in a specified step length, an on-vehicle safety worker gives an early warning to the on-vehicle safety worker at the moment, the on-site intervention of the safety worker ensures the safety of passengers and vehicles, and if the on-vehicle safety worker does not exist, an early warning signal is uploaded to a remote monitoring center to be subjected to remote panoramic intervention by the safety worker or a safety plan is automatically taken to ensure the safety of the passengers and vehicles.
The scheme has the following effects:
(1) the road-end-vehicle-end cooperative data acquisition advantages are fully exerted, and the driving data which is wide in coverage sample, various in scene, large in size and mutually complementary in characteristic can be efficiently acquired;
(2) establishing an objective evaluation system of the automatic driving comfort degree of the automatic driving vehicle under different speeds and different curvatures, wherein the objective evaluation system can evaluate the driving comfort degree of the automatic driving vehicle on line according to the objective evaluation system, and the obtained reference standard value can be used as a state constraint during development of an automatic driving planning control algorithm to guide more reasonable track development;
(3) the neural network is trained based on massive data, abnormal behaviors of the vehicle and vehicles around a certain range can be predicted on line, when the abnormal behaviors occur, early warning is timely carried out to a vehicle-mounted safety worker or a remote monitoring center, and intervention measures are timely taken to ensure safety of passengers and the vehicles.
In addition to the above embodiments, the present invention may have other embodiments. All technical solutions formed by adopting equivalent substitutions or equivalent transformations fall within the protection scope of the claims of the present invention.
Claims (10)
1. A method for evaluating and early warning states of an automatic driving vehicle based on vehicle-road cooperation is characterized by comprising the following steps: the method comprises the following steps:
s1 data acquisition based on vehicle-road cooperation technology
S1.1, for a road side, acquiring driving data of an automatic driving vehicle and a non-automatic driving vehicle passing through the road section through a sensing system of a road side unit, and uploading the driving data to a cloud end platform through a road side unit network layer;
s1.2, for the side of the vehicle, the automatic driving vehicle is driven by a driver, vehicle driving data are collected by means of a vehicle-mounted sensing layer, and the driving data are transmitted to a cloud platform through a vehicle-mounted network terminal;
s2, somatosensory comfort state assessment
S2.1 somatosensory comfort evaluation index: accx: longitudinal acceleration, m/s2(ii) a decx: longitudinal deceleration, m/s2(ii) a accy: lateral acceleration, m/s2(ii) a jerkx: longitudinal jerk, m/s3(ii) a jerky: transverse jerk, m/s3；
S2.2, calculating the somatosensory comfort data;
s2.3, obtaining comfort index standard reference values under different speeds and curvatures according to different rules;
s2.4, on-line evaluation of somatosensory comfort: designing a hysteresis logic and a hysteresis curved surface, comparing the somatosensory data value with a standard somatosensory data hysteresis interval value in an online mode, and outputting the specific type of the uncomfortable state, the frequency of the uncomfortable state in the process of one-time automatic driving and the proportion of the uncomfortable time to the total driving time;
s3, safety state evaluation and early warning
A large amount of data covered by multiple scenes acquired by a vehicle-road cooperation technology is used as a data set for offline training of a neural network model, the model is output to judge whether abnormal behaviors occur in a vehicle within a period of time in the future, and the model is used for online prediction of self-driven vehicles and social vehicles within a certain range.
2. The automatic driving vehicle state assessment and early warning method based on vehicle-road coordination as claimed in claim 1, wherein: in the S1.1, the driving data comprises a plurality of items of information including a vehicle coordinate point position, a longitudinal speed and an acceleration, a transverse speed and an acceleration, a vehicle course angle, a road curvature and a steering radius under a global coordinate system.
3. The automatic driving vehicle state assessment and early warning method based on vehicle-road coordination as claimed in claim 1, wherein: in S1.1, the perception system includes laser radar, camera, millimeter wave radar, in S1.2, on-vehicle perception layer includes IMU, laser radar, millimeter wave radar.
4. The automatic driving vehicle state assessment and early warning method based on vehicle-road coordination as claimed in claim 1, wherein: s2.2 somatosensory comfort data calculation:
accx: vehicle end: calculating according to the position coordinate information difference to obtain longitudinal acceleration, performing second-order low-pass filtering to obtain est _ accx, performing second-order low-pass filtering to the acceleration information obtained based on the IMU to obtain IMU _ accx, adding a confidence ratio to the two acceleration values to obtain ax _ filtered Ka est _ accx + Kb IMU _ accx, and performing zero-crossing detection to obtain a final longitudinal acceleration value accx; road end: according to the position coordinate information, carrying out differential calculation to obtain longitudinal acceleration, carrying out second-order low-pass filtering to obtain est _ accx, and carrying out zero-crossing detection to obtain a final longitudinal acceleration value accx;
decx: the method is consistent with the accx calculation method, and a final longitudinal acceleration value decx is obtained through zero-crossing detection;
accy: calculating to obtain curvature information kappa of the current running road as dyaw/ds according to the position information, wherein dyaw is the change of a course angle in unit time, and ds is the change of a vehicle position in unit time; vehicle end: performing first-order low-pass filtering on the lateral acceleration data based on the IMU to obtain IMU _ accy; road end: calculating est _ acy by using a position difference method;
jerkx: carrying out difference based on the longitudinal acceleration to obtain a longitudinal acceleration degree change rate jerk, and carrying out mean value filtering to obtain est _ jerkx;
jerky: and carrying out difference based on the lateral acceleration to obtain lateral acceleration change rate jerk, and carrying out robust mean filtering to obtain est _ jerky.
5. The automatic driving vehicle state assessment and early warning method based on vehicle-road coordination as claimed in claim 1, wherein: s2.3 somatosensory comfort evaluation standard:
(1) processing vehicle driving data obtained by the vehicle road cooperation, calculating the curvature of a vehicle driving route and the current vehicle speed at each moment, and obtaining various driving comfort index values under the curvature-vehicle speed, so that a huge data lattice set is formed, and five kinds of curvature-speed-comfort index scattered point distributions can be obtained by separating and processing the data lattice set;
(2) defining a speed anchor point and a curvature anchor point, wherein the minimum turning diameter of the passenger car is 9.0-12.0m, and designing the curvature interval to be 0-0.2m-1The interval size is 0.02, namely [0:0.02:0.2 ]]The designed vehicle speed interval is 0-22m/s, and the interval size is 1.0, namely [0:1:22 ]]；
(3) Screening out bad data, traversing speed anchor points and curvature anchor points, calculating extreme values, average values, mean square deviations, 25%, 50% and 75% data value distribution points of various comfort indexes in the range of anchor points, and fitting the data by using a mobile least square method;
(4) comparing and analyzing the vehicle end data, the road end data and the vehicle road cooperative data;
(5) and obtaining comfort index standard reference values under different speeds and curvatures according to different rules.
6. The automatic driving vehicle state assessment and early warning method based on vehicle-road coordination as claimed in claim 5, wherein: and S2.3(5), designing an evaluation function, or modifying the penalty coefficient to obtain a proper reference standard value, or taking the Gaussian function as a membership function and then carrying out clarification processing to obtain the reference standard value.
7. The automatic driving vehicle state evaluation and early warning method based on vehicle-road cooperation as claimed in claim 1, characterized in that: and S2.3, selecting 75% of numerical distribution points as comfort reference standard values.
8. The automatic driving vehicle state assessment and early warning method based on vehicle-road coordination as claimed in claim 1, wherein: and S2.4, performing interpolation calculation by adopting a curved surface interpolation method to obtain the hysteresis upper limit and the hysteresis lower limit of the standard somatosensory data under the current kappa and speed, and performing online evaluation on the somatosensory comfort level according to the obtained standard interval and logic.
9. The automatic driving vehicle state assessment and early warning method based on vehicle-road coordination as claimed in claim 1, wherein: and S2, obtaining the total comfort evaluation comparison grade corresponding to the discomfort times and the discomfort time proportion according to a fuzzy statistical method, wherein the steps are as follows:
(1) the automatic driving time is less than or equal to 10mins
The discomfort times are less than or equal to 6 times, the discomfort time proportion is less than or equal to 10 percent, and the comfort level is high;
the discomfort times are less than or equal to 18 times, the discomfort time proportion is less than or equal to 25 percent, and the comfort level is medium;
the discomfort times are less than or equal to 30 times, the discomfort time proportion is less than or equal to 40 percent, and the comfort level is low;
(2) the automatic driving time is less than or equal to 20mins
The discomfort times are less than or equal to 15 times, the discomfort time proportion is less than or equal to 15 percent, and the comfort level is high;
the discomfort times are less than or equal to 45 times, the discomfort time proportion is less than or equal to 30 percent, and the comfort level is medium;
the discomfort times are less than or equal to 80, the discomfort time proportion is less than or equal to 45 percent, and the comfort level is low;
(3) the automatic driving time is less than or equal to 40mins
The discomfort times are less than or equal to 45 times, the discomfort time proportion is less than or equal to 20 percent, and the comfort level is high;
the discomfort times are less than or equal to 150 times, the discomfort time proportion is less than or equal to 35 percent, and the comfort level is medium;
the uncomfortable times are less than or equal to 300 times, the proportion of the uncomfortable time is less than or equal to 50 percent, and the comfort level is low.
10. The automatic driving vehicle state assessment and early warning method based on vehicle-road coordination as claimed in claim 1, wherein: and S3, the abnormal behaviors comprise abnormal parking, abnormal acceleration, abnormal deceleration, abnormal large-amplitude steering wheel rotation and collision risk.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111675077.5A CN114550442B (en) | 2021-12-31 | 2021-12-31 | Automatic driving vehicle state evaluation and early warning method based on vehicle-road cooperation |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111675077.5A CN114550442B (en) | 2021-12-31 | 2021-12-31 | Automatic driving vehicle state evaluation and early warning method based on vehicle-road cooperation |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114550442A true CN114550442A (en) | 2022-05-27 |
| CN114550442B CN114550442B (en) | 2023-11-03 |
Family
ID=81669147
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111675077.5A Active CN114550442B (en) | 2021-12-31 | 2021-12-31 | Automatic driving vehicle state evaluation and early warning method based on vehicle-road cooperation |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114550442B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115862188A (en) * | 2022-12-01 | 2023-03-28 | 西部科学城智能网联汽车创新中心(重庆)有限公司 | Method and device for sensing abnormal driving based on cloud |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106503378A (en) * | 2016-10-26 | 2017-03-15 | 哈尔滨工业大学 | A kind of flat vertical line shape Evaluation Method of Mechanical Property of the road based on transverse acceleration and vertical acceleration |
| CN107807625A (en) * | 2017-09-05 | 2018-03-16 | 百度在线网络技术（北京）有限公司 | A kind of appraisal procedure and device based on automated driving system comfort level end to end |
| CN108248608A (en) * | 2017-12-28 | 2018-07-06 | 北京百度网讯科技有限公司 | For assessing the method and apparatus of the comfort level of control loop |
| CN109583776A (en) * | 2018-12-05 | 2019-04-05 | 百度在线网络技术（北京）有限公司 | A kind of vehicle body-sensing evaluating method, device, electronic equipment, medium and vehicle |
| CN111540237A (en) * | 2020-05-19 | 2020-08-14 | 河北德冠隆电子科技有限公司 | Method for automatically generating vehicle safety driving guarantee scheme based on multi-data fusion |
| CN111829793A (en) * | 2020-08-03 | 2020-10-27 | 广州导远电子科技有限公司 | Driving process comfort evaluation method, device and system based on combined positioning |
| JP2020175893A (en) * | 2015-11-04 | 2020-10-29 | トヨタ自動車株式会社 | Automated driving system, automated driving apparatus, and automated driving method |
| CN112418646A (en) * | 2020-11-18 | 2021-02-26 | 广州小鹏自动驾驶科技有限公司 | Vehicle comfort evaluation method and device and readable storage medium |
| CN113370984A (en) * | 2021-06-30 | 2021-09-10 | 中国科学技术大学先进技术研究院 | Multi-index-based comprehensive evaluation method and system for comfort of automatic driving vehicle |
-
2021
- 2021-12-31 CN CN202111675077.5A patent/CN114550442B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2020175893A (en) * | 2015-11-04 | 2020-10-29 | トヨタ自動車株式会社 | Automated driving system, automated driving apparatus, and automated driving method |
| CN106503378A (en) * | 2016-10-26 | 2017-03-15 | 哈尔滨工业大学 | A kind of flat vertical line shape Evaluation Method of Mechanical Property of the road based on transverse acceleration and vertical acceleration |
| CN107807625A (en) * | 2017-09-05 | 2018-03-16 | 百度在线网络技术（北京）有限公司 | A kind of appraisal procedure and device based on automated driving system comfort level end to end |
| CN108248608A (en) * | 2017-12-28 | 2018-07-06 | 北京百度网讯科技有限公司 | For assessing the method and apparatus of the comfort level of control loop |
| CN109583776A (en) * | 2018-12-05 | 2019-04-05 | 百度在线网络技术（北京）有限公司 | A kind of vehicle body-sensing evaluating method, device, electronic equipment, medium and vehicle |
| CN111540237A (en) * | 2020-05-19 | 2020-08-14 | 河北德冠隆电子科技有限公司 | Method for automatically generating vehicle safety driving guarantee scheme based on multi-data fusion |
| CN111829793A (en) * | 2020-08-03 | 2020-10-27 | 广州导远电子科技有限公司 | Driving process comfort evaluation method, device and system based on combined positioning |
| CN112418646A (en) * | 2020-11-18 | 2021-02-26 | 广州小鹏自动驾驶科技有限公司 | Vehicle comfort evaluation method and device and readable storage medium |
| CN113370984A (en) * | 2021-06-30 | 2021-09-10 | 中国科学技术大学先进技术研究院 | Multi-index-based comprehensive evaluation method and system for comfort of automatic driving vehicle |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115862188A (en) * | 2022-12-01 | 2023-03-28 | 西部科学城智能网联汽车创新中心(重庆)有限公司 | Method and device for sensing abnormal driving based on cloud |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114550442B (en) | 2023-11-03 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109177982B (en) | Vehicle driving risk degree evaluation method considering driving style | |
| CN109767651B (en) | Seamless communication method for curve area in V2X environment | |
| DE102017103970A1 (en) | Transition between autonomous vehicle control levels | |
| DE102017103973A1 (en) | Autonomous vehicle control taking into account the driver's attention | |
| DE102017103969A1 (en) | Autonomous vehicle control with risk factor | |
| CN106373331A (en) | Riding early-warning method and device | |
| CN106952361B (en) | A kind of efficient vehicle running state identifying system | |
| DE102009034097A1 (en) | Adaptive vehicle control system with integrated driving style recognition | |
| CN105030257B (en) | Driver's abnormality detection | |
| DE102009034096A1 (en) | Adaptive vehicle control system with driving style recognition | |
| CN110660194A (en) | Driving monitoring early warning method and system | |
| DE102017117240A1 (en) | INFORMATION ANSWER FOR A VEHICLE | |
| CN113895464B (en) | Intelligent vehicle driving map generation method and system fusing personalized driving style | |
| CN112950811B (en) | New energy automobile region operation risk assessment and early warning system integrating whole automobile safety | |
| CN110428621A (en) | A kind of monitoring of Floating Car dangerous driving behavior and method for early warning based on track data | |
| CN110696835B (en) | Automatic early warning method and automatic early warning system for dangerous driving behaviors of vehicle | |
| CN114550442B (en) | Automatic driving vehicle state evaluation and early warning method based on vehicle-road cooperation | |
| CN115620530A (en) | Automatic driving takeover risk assessment model considering driving risk field and construction method thereof | |
| CN110320916A (en) | Consider the autonomous driving vehicle method for planning track and system of occupant's impression | |
| CN116572984A (en) | Dangerous driving management and control method and system based on multi-feature fusion | |
| CN112070927A (en) | Highway vehicle microscopic driving behavior analysis system and analysis method | |
| CN115892063A (en) | Road condition monitoring and coping method for unmanned commercial vehicle | |
| CN113335293B (en) | Highway road surface detection system of drive-by-wire chassis | |
| CN114987539A (en) | Individual collision grading early warning method and system for automatic driving automobile based on risk field model | |
| CN113240901B (en) | Intelligent vehicle lane changing behavior risk level determination method and device |
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 | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20230418 Address after: 1 / F, building 28, 6055 Jinhai highway, Fengxian District, Shanghai 201400 Applicant after: Zhongzhixing (Shanghai) Transportation Technology Co.,Ltd. Applicant after: Tianyi Transportation Technology Co.,Ltd. Address before: 1 / F, building 28, 6055 Jinhai highway, Fengxian District, Shanghai 201400 Applicant before: Zhongzhixing (Shanghai) Transportation Technology Co.,Ltd. |
|
| TA01 | Transfer of patent application right | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |