CN115620530A - Automatic driving takeover risk assessment model considering driving risk field and construction method thereof - Google Patents
Automatic driving takeover risk assessment model considering driving risk field and construction method thereof Download PDFInfo
- Publication number
- CN115620530A CN115620530A CN202211137773.5A CN202211137773A CN115620530A CN 115620530 A CN115620530 A CN 115620530A CN 202211137773 A CN202211137773 A CN 202211137773A CN 115620530 A CN115620530 A CN 115620530A
- Authority
- CN
- China
- Prior art keywords
- vehicle
- risk
- driving
- assessment model
- takeover
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000012502 risk assessment Methods 0.000 title claims abstract description 37
- 238000010276 construction Methods 0.000 title claims abstract description 9
- 238000000034 method Methods 0.000 claims abstract description 57
- 238000005315 distribution function Methods 0.000 claims abstract description 30
- 230000001133 acceleration Effects 0.000 claims abstract description 25
- 238000001914 filtration Methods 0.000 claims abstract description 15
- 230000003068 static effect Effects 0.000 claims abstract description 15
- 230000035484 reaction time Effects 0.000 claims abstract description 11
- 230000004044 response Effects 0.000 claims abstract description 11
- 230000002452 interceptive effect Effects 0.000 claims abstract description 9
- 238000007781 pre-processing Methods 0.000 claims abstract description 4
- 239000011159 matrix material Substances 0.000 claims description 48
- 230000008569 process Effects 0.000 claims description 31
- 238000004364 calculation method Methods 0.000 claims description 19
- 238000004422 calculation algorithm Methods 0.000 claims description 12
- 238000013210 evaluation model Methods 0.000 claims description 12
- 230000008859 change Effects 0.000 claims description 9
- 230000014509 gene expression Effects 0.000 claims description 9
- 150000001875 compounds Chemical class 0.000 claims description 7
- 230000002068 genetic effect Effects 0.000 claims description 6
- 229940057373 c-time Drugs 0.000 claims description 3
- 230000035772 mutation Effects 0.000 claims description 3
- 230000009467 reduction Effects 0.000 claims description 3
- 238000005070 sampling Methods 0.000 claims description 3
- 238000011160 research Methods 0.000 description 4
- 230000000694 effects Effects 0.000 description 2
- 238000011156 evaluation Methods 0.000 description 2
- 238000011158 quantitative evaluation Methods 0.000 description 2
- 238000004088 simulation Methods 0.000 description 2
- 206010039203 Road traffic accident Diseases 0.000 description 1
- 230000002159 abnormal effect Effects 0.000 description 1
- 230000002457 bidirectional effect Effects 0.000 description 1
- 239000002131 composite material Substances 0.000 description 1
- 238000012937 correction Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 230000008447 perception Effects 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 230000000007 visual effect Effects 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
- G08G1/052—Detecting movement of traffic to be counted or controlled with provision for determining speed or overspeed
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/12—Computing arrangements based on biological models using genetic models
- G06N3/126—Evolutionary algorithms, e.g. genetic algorithms or genetic programming
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0635—Risk analysis of enterprise or organisation activities
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/10—Services
- G06Q50/26—Government or public services
- G06Q50/265—Personal security, identity or safety
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/09—Arrangements for giving variable traffic instructions
- G08G1/0962—Arrangements for giving variable traffic instructions having an indicator mounted inside the vehicle, e.g. giving voice messages
- G08G1/0967—Systems involving transmission of highway information, e.g. weather, speed limits
- G08G1/096708—Systems involving transmission of highway information, e.g. weather, speed limits where the received information might be used to generate an automatic action on the vehicle control
- G08G1/096725—Systems involving transmission of highway information, e.g. weather, speed limits where the received information might be used to generate an automatic action on the vehicle control where the received information generates an automatic action on the vehicle control
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Human Resources & Organizations (AREA)
- Life Sciences & Earth Sciences (AREA)
- Health & Medical Sciences (AREA)
- Theoretical Computer Science (AREA)
- Strategic Management (AREA)
- Biophysics (AREA)
- Economics (AREA)
- Tourism & Hospitality (AREA)
- Entrepreneurship & Innovation (AREA)
- General Health & Medical Sciences (AREA)
- Evolutionary Biology (AREA)
- Marketing (AREA)
- Bioinformatics & Computational Biology (AREA)
- General Business, Economics & Management (AREA)
- Educational Administration (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Development Economics (AREA)
- Atmospheric Sciences (AREA)
- Biomedical Technology (AREA)
- Computer Security & Cryptography (AREA)
- Game Theory and Decision Science (AREA)
- Quality & Reliability (AREA)
- Operations Research (AREA)
- Physiology (AREA)
- Genetics & Genomics (AREA)
- Artificial Intelligence (AREA)
- Primary Health Care (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Traffic Control Systems (AREA)
Abstract
The invention relates to a construction method of an automatic driving takeover risk assessment model considering a driving risk field, which comprises the following steps: step one, calculating a vehicle performance probability factor according to the response speed and the response degree of a vehicle to risks; step two, constructing a distance-based static risk distribution function; step three, constructing a dynamic risk distribution function based on the motion state; step four, considering the influence of the take-over reaction time, and calculating the field intensity of the driving risk field; and extracting data information from the vehicle and the interactive vehicle; step five, preprocessing the speed and acceleration data of the vehicle based on a Kalman filtering method; and step six, substituting the calibrated equivalent mass and various to-be-fixed constant coefficients into the automatic driving takeover risk assessment model to obtain the automatic driving takeover risk assessment model based on the driving risk field. The invention can improve the safety of the whole road network system and reduce the accident rate and the casualty rate caused by automatically driving vehicles.
Description
Technical Field
The invention belongs to the technical field of traffic safety, and particularly relates to an automatic driving takeover risk assessment model considering a driving risk field and a construction method thereof.
Background
When a vehicle in an automatic driving state faces dangerous and emergency road conditions beyond the capability range of the vehicle or is in a failure and fault state, a driver needs to perform emergency treatment to take over the vehicle in an emergency. Since the driver often engages in a secondary task unrelated to driving in the automatic driving state, the awareness and reaction to the traffic environment are reduced to various degrees, so that the risk response time of the driver in the automatic driving state is longer than that in the manual driving state, and traffic conflicts or accidents are easily caused.
The existing automatic driving takeover research is mainly developed aiming at the influence indexes and change rules of takeover performance, quantitative evaluation research on takeover risks needs to be further carried out, and the existing mainstream risk evaluation model is mainly aimed at non-automatic driving vehicles and the traditional following lane changing scene, so that the applicable driving scene is single. Therefore, from the perspective of vehicle dynamic parameters, considering driving risk field characteristics, an automatic driving takeover risk assessment model is urgently needed at present, and a method and technical support can be provided for takeover risk assessment and driving assistance system improvement under different driving environments.
Disclosure of Invention
The invention aims to solve the technical problems and further provides an automatic driving takeover risk assessment model considering a driving risk field and a construction method thereof.
The invention relates to a construction method of an automatic driving takeover risk assessment model considering a driving risk field, which comprises the following steps:
step one, calculating a vehicle performance probability factor according to the response speed and the response degree of a vehicle to risks;
step two, constructing a distance-based static risk distribution function;
step three, constructing a dynamic risk distribution function based on the motion state;
step four, considering the influence of the take-over reaction time, and calculating the field intensity of the driving risk field; and extracting data information from the vehicle and the interactive vehicle;
step five, preprocessing the speed and acceleration data of the vehicle based on a Kalman filtering method;
step six, calibrating the equivalent quality of the takeover risk evaluation model; calibrating each to-be-calibrated coefficient of the connecting pipe risk evaluation model; and substituting the calibrated equivalent mass and various to-be-fixed constant coefficients into the automatic driving takeover risk assessment model to obtain the automatic driving takeover risk assessment model based on the driving risk field.
In the first step, a calculation formula of the vehicle performance probability factor is as follows:
where ζ represents a coefficient of state of the vehicle, ζ =1 for a normal vehicle, and 0 for a faulty vehicle<ζ<1; c-the magnitude (N) of the depression of the brake pedal of the vehicle in the take-over process; c max -maximum value of the depression amplitude (N) of the brake pedal of the vehicle during take-over; z-the steering wheel angle (°) of the vehicle during take over; z max -maximum steering wheel angle (°) of the vehicle during take over; alpha is alpha 1 ，α 2 To be constant, α 1 >0，α 2 >0。
In the second step, the distance-based static risk distribution function calculation formula is as follows:
in the formula, M i -equivalent mass (kg) of the vehicle i; kappa-coefficient to be constant, kappa>0；d ji -the distance (m) between a point j in the road plane and the own vehicle i; gamma, the descent rate factor, can take the value of 1.
3. The method for constructing the risk assessment model taking over the automatic driving under consideration of the driving risk field according to claim 1, wherein in the third step, the dynamic risk distribution function calculation formula based on the motion state is as follows:
in the formula, M i -equivalent mass (kg) of the own vehicle i; a-resultant deceleration (m/s) of the own vehicle i during take-over 2 ) (ii) a Delta-the resultant deceleration reduction factor, delta>0, can take the value of 0.15; theta ij -the angle theta between the connecting line direction of the bicycle i and other traffic participants j in the driving risk field and the positive direction of the x axis in the pipe taking process ij ∈[0°,180°]。
In the fourth step, the calculation formula of the driving risk field intensity is as follows:
in the formula: s i The driving risk field intensity of the self vehicle i;
the dynamic risk distribution function generated by the traffic unit j at the self vehicle i in the takeover process;
the static risk distribution function generated by the traffic unit j at the self vehicle i in the takeover process; p is c -a vehicle performance probability factor; t is t i -take-over reaction time(s) of the vehicle i in the take-over process; lambda 1 -the dynamic risk distribution function accounts for weight; lambda 2 -the weight taken up by the static risk distribution function; Φ — undetermined constant coefficient;
the take over risk index is calculated as follows:
in the formula (I), the compound is shown in the specification,RISK i -take over risk index of own car i;
-an average value of the driving risk field strength;
substituting expressions (1) to (3) into expression (4) and combining expression (5) to obtain an automatic driving takeover risk evaluation model considering a driving risk field, wherein the automatic driving takeover risk evaluation model comprises the following steps:
5. the method for constructing the automatic driving takeover risk assessment model considering the driving risk field according to claim 1, wherein in the fourth step, the real-time coordinate position, the linear acceleration, the driving speed, the steering wheel angle, the brake pedal opening degree and the takeover reaction time of the driver of the vehicle are extracted; data information from and to the interactive vehicle including the type of interactive vehicle, real-time location coordinates, and speed of the vehicle.
In the fifth step, the specific algorithm flow is as follows:
(1) Calculating a state prediction value of a Kalman filtering algorithm:
in the formula (I), the compound is shown in the specification,
-observed variables of the speed and acceleration of the vehicle,
v r -speed of vehicle filtering (m/s); a is r -vehicle filtered acceleration (m/s) -2 ) (ii) a E-coefficient matrix of system prediction state equation; f-input control item matrix; g is the control quantity of the current state; kappa r -a priori predicting process noise in the process; xi r-1 Time of accelerationRate of change between (m/s) -3 )；
The coefficient matrix E of the system prediction state equation, the input control item matrix F and the calculation formula of the control quantity G of the current state are shown as the formula:
where T is the inverse(s) of the data sampling frequency.
(2) Calculating a filter gain matrix
Assumption xi r-1 And kappa r All satisfy normal distribution, and the covariance is J and C, respectively, then the filter gain matrix is formula (9):
in the formula (I), the compound is shown in the specification,
-covariance matrix of a priori estimated errors (observations); f-input control item matrix; e-coefficient matrix of system prediction state equation; C-Process noise in Prior prediction Process κ r The covariance of (a).
(3) The covariance matrix of the a priori estimation errors (observations) is given by equation (3-23):
in the formula, F is the control item matrix of the input;
-observed variables of the speed and acceleration of the vehicle,
g is the control quantity of the current state; c-time rate of change xi of acceleration r-1 Assistant ofAnd (4) poor.
(4) The covariance matrix of the a posteriori estimation errors (predicted values) is given by equation (3-23):
wherein, K is an identity matrix; w r -a filter gain matrix; e-coefficient matrix of system prediction state equation;
-covariance matrix of a priori estimated errors (observations).
In the sixth step, the equivalent mass is calibrated based on the actual mass, type and speed of the vehicle, and the calculation formula is as follows:
M i ＝M(m i ,v i )＝m i ·(ρ·v i u +χ) (12)
in the formula, m i -actual mass (kg) of the own vehicle i; v. of i -the speed of travel (m/s) of the own vehicle i; ρ, u, χ -undetermined constant coefficient.
In the sixth step, each to-be-constant coefficient of the takeover risk assessment model is calibrated by adopting a genetic algorithm, and the genetic parameters are configured as follows: setting the size and scale of a population to be 200; the crossover probability is 0.8; the mutation probability is 0.2, and the iteration times are 300.
The invention also relates to a model utilizing the construction method of the automatic driving takeover risk assessment model considering the driving risk field.
Advantageous effects
The method takes the quantitative evaluation of the risk existing in the take-over process of the L3-level automatic driving vehicle as a starting point, takes the driving risk field thought as a theoretical basis, introduces the vehicle performance probability factor and considers the influence of take-over reaction time by establishing a distance-based static risk distribution function and a motion state-based dynamic risk distribution function, and constructs the automatic driving take-over risk evaluation model considering the driving risk field.
The method and the system can provide theoretical basis and reference for the taking-over risk research and safety evaluation of the automatic driving vehicle under the intelligent traffic system, and also provide a new research idea for the taking-over strategy design of the automatic driving vehicle, thereby improving the safety of the whole road network system and reducing the accident rate and the casualty rate caused by the automatic driving vehicle.
Drawings
Fig. 1 is a flowchart of a method for constructing an automatic driving takeover risk assessment model in consideration of a driving risk field according to the present invention.
Detailed Description
The present embodiment will be specifically described below with reference to fig. 1.
The invention provides a construction method of an automatic driving takeover risk assessment model considering a driving risk field, which comprises the following steps:
the method comprises the following steps: the method comprises the steps of (1) introducing a vehicle performance probability factor by considering the response speed and the response degree of a vehicle to risks;
in the process of taking over during driving, the vehicle has two main ways of avoiding risks: braking and steering. The response speed and the response degree of the vehicle to the risk in the process of taking over are measured by the depression amplitude of the brake pedal and the steering wheel angle of the vehicle. The smaller the pressing amplitude of the brake pedal is, the lower the control degree of the driver on the vehicle is, and the more easily the collision danger is generated; the greater the steering wheel angle of the vehicle, the greater the likelihood that the driver will successfully bypass the obstacle when taking over. The calculation formula of the vehicle performance probability factor is as follows:
where ζ represents a coefficient of state of the vehicle, ζ =1 for a normal vehicle, and 0 for a faulty vehicle<ζ<1; c-the magnitude (N) of the depression of the brake pedal of the vehicle in the take-over process; c max -maximum value of the depression amplitude (N) of the brake pedal of the vehicle during take-over; z-the steering wheel angle (°) of the vehicle during take-over; z is a linear or branched member max -a maximum steering wheel angle (°) of the vehicle during take over; alpha is alpha 1 ，α 2 To be constant, α 1 >0，α 2 >0。
Step two: considering the distance between the vehicle and other traffic units in the taking over process, the variation trend of the distance and the equivalent mass of the vehicle, and constructing a distance-based static risk distribution function;
another factor that influences the risk of take-over is the distance between this and other traffic units and the trend of the distance change during take-over, due to the limitations of the driver's perception ability. In road traffic, the greater the distance from the driving risk field source, the smaller the probability of a collision accident, i.e., the weaker the field strength of the driving risk field generated by the field source. In a driving risk field, a distance-based static risk distribution function calculation formula is as follows:
in the formula, M i -equivalent mass (kg) of the own vehicle i; kappa-undetermined constant coefficient, kappa>0；d ji -the distance (m) between a point j in the road plane and the own vehicle i; gamma, the descent rate factor, can take the value of 1.
Step three: considering the motion states of the own vehicle and other surrounding traffic participants and the equivalent mass of the own vehicle in the takeover process, and constructing a dynamic risk distribution function based on the motion states;
the greater the moving speed of the own vehicle, the greater the field intensity generated by the traffic unit in a moving state around the own vehicle. When the vehicle encounters an obstacle to perform deceleration movement, the smaller the deceleration value is, the slower the speed change is, the poorer the braking effect is, and the higher the probability of a collision accident is. Further, the smaller the angle between the direction vector of the host vehicle directed to the surrounding traffic participants and the x-axis forward direction (defined as the road direction and coinciding with the direction in which the host vehicle is heading), the greater the risk of traveling. The dynamic risk distribution function calculation formula of the traffic unit based on the motion state in the take-over process is as follows:
in the formula, M i -equivalent mass (kg) of the vehicle i; a-resultant deceleration (m/s) of the own vehicle i during take-over 2 ) (ii) a Delta-the resultant deceleration reduction factor, delta>0, can take the value of 0.15; theta.theta. ij -the angle theta between the direction of the connection line of the vehicle i and the other traffic participants j in the driving risk field and the positive direction of the x axis in the taking over process ij ∈[0°,180°]。
Step four: introducing a vehicle performance probability factor to represent the probability of occurrence of a potential traffic accident caused by abnormal vehicle track in the takeover process, describing the distribution characteristics of the potential risk of the vehicle in the traffic scene in the driving takeover process by adopting a risk distribution function, and considering the influence of takeover reaction time, wherein the calculation formula of the driving risk field strength is as follows:
in the formula: s i The driving risk field intensity of the own vehicle i;
-a dynamic risk distribution function generated by the traffic unit j at the host vehicle i during take-over;
-a static risk distribution function generated by the traffic unit j at the host vehicle i during take-over; p c -a vehicle performance probability factor; t is t i -take-over reaction time(s) of the vehicle i in the take-over process; lambda [ alpha ] 1 -the dynamic risk distribution function accounts for weight; lambda [ alpha ] 2 -the weight taken up by the static risk distribution function; phi-undetermined constant coefficient.
The driving risk field intensity is an absolute index, the numerical variation range is large, and the risk in the process of taking over is difficult to effectively evaluate. Therefore, the index needs to be processed, and the relative index is defined as the "take-over risk index" so as to evaluate the risk in the take-over process. The larger the take-over risk index, the more likely the vehicle is to be at risk during take-over at that point. The take over risk index is calculated as follows:
in the formula, RISK i -take over risk index of own car i;
Substituting expressions (1) to (3) into expression (4) and combining expression (5) to obtain an automatic driving takeover risk assessment model considering a driving risk field as follows:
step five: extracting data information from the vehicle and the interactive vehicle;
extracting real-time coordinate position, linear acceleration (x/y/z direction), driving speed, steering wheel angle, brake pedal opening and take-over reaction time of a driver of the vehicle so as to calculate deceleration, inverse collision time and driving risk field strength of a subsequent composite vehicle;
the type of interactive vehicle, the real-time position coordinates and the speed of the vehicle are extracted for subsequent calculation of the dynamic and static risk distribution functions.
Step six: preprocessing speed and acceleration data of the vehicle based on a Kalman filtering method;
the method is characterized in that the speed and the acceleration of a vehicle are processed in time by adopting a Kalman filtering algorithm, the Kalman filtering algorithm is an algorithm taking a state equation as a core, and the principle is to take the distribution condition of data at different time points into consideration and fuse the data. The standard kalman filter is divided into two aspects: the first is prior forecast, and the second is forecast and correction. The specific algorithm flow is as follows:
(1) Calculating a state prediction value of a Kalman filtering algorithm:
in the formula (I), the compound is shown in the specification,
-the observed variables of the speed and acceleration of the vehicle,
v r -the speed of vehicle filtering (m/s); a is a r -vehicle filtered acceleration (m/s) -2 ) (ii) a E-coefficient matrix of system prediction state equation; f-input control item matrix; g is the control quantity of the current state; kappa r -a priori predicting process noise in the process; xi r-1 Time rate of change of acceleration (m/s) -3 )。
The calculation formula of the coefficient matrix E of the system prediction state equation, the input control item matrix F and the control quantity G of the current state is shown as the formula (8):
where T is the inverse(s) of the data sampling frequency.
(2) Calculating a filter gain matrix
Assumption xi r-1 And kappa r All satisfy normal distribution, and the covariance is J and C, respectively, then the filter gain matrix is formula (3-21):
in the formula (I), the compound is shown in the specification,
-covariance matrix of a priori estimated errors (observations); f-input control item matrix; e-coefficient matrix of system prediction state equation; C-Process noise in Prior prediction Process κ r The covariance of (a).
(3) The covariance matrix of the a priori estimation errors (observations) is given by equation (3-23):
in the formula, F is the control item matrix of the input;
-observed variables of the speed and acceleration of the vehicle,
g is the control quantity of the current state; c-time rate of change xi of acceleration r-1 The covariance of (a).
(4) The covariance matrix of the a posteriori estimation error (predicted value) is the formula (3-23):
wherein, K is an identity matrix; w r -a filter gain matrix; e-coefficient matrix of system prediction state equation;
-covariance matrix of a priori estimated errors (observations).
Step seven: calibrating the equivalent quality of the takeover risk evaluation model;
the equivalent mass represents the risk of the vehicle attribute to the vehicle, and the larger the equivalent mass is, the greater the severity of the collision of the vehicle with other traffic participants. The equivalent mass is not only related to the actual mass of the vehicle, but also to its type and speed. Therefore, the equivalent mass is calibrated based on the actual mass, type and speed of the vehicle, and the calculation formula is as follows:
M i ＝M(m i ,v i )＝m i ·(ρ·v i u +χ) (12)
in the formula, m i -actual mass (kg) of the vehicle i; v. of i -the speed of travel (m/s) of the host vehicle i; ρ, u, χ, undetermined constant coefficient.
Step eight: calibrating each to-be-calibrated coefficient of the connecting pipe risk evaluation model;
and (3) calibrating each to-be-calibrated coefficient of the take-over risk evaluation model by adopting a genetic algorithm, wherein the genetic parameters are configured as follows: setting the size and scale of a population to be 200; the crossover probability is 0.8; the mutation probability is 0.2, and the iteration times are 300.
Step nine: and (6) substituting the calibrated equivalent mass and various coefficients to be stabilized into a formula (6), thus obtaining the automatic driving takeover risk assessment model based on the driving risk field.
Examples
And selecting the root mean square error as an evaluation index, and verifying the effectiveness of the automatic driving takeover Risk evaluation model considering the driving Risk field by comparing the reciprocal of the collision time with the Risk value of the constructed model.
The reciprocal of the collision time is a common index for evaluating the driving risk of the vehicle, and is calculated as follows:
in the formula, TD i Inverse collision time(s) of the host vehicle i -1 )；v i -the speed of travel (m/s) of the host vehicle i during braking; v. of j During braking, the speed of travel of the preceding vehicle j (m/s, v if the front is a stationary obstacle j ＝0)；d ij -the distance (m) between the own vehicle i and the preceding vehicle j. And collecting related data by using a driving simulation method to verify the effectiveness.
The simulation scene is constructed based on SCANeR studio software, the road type is a straight line section of an urban expressway, the total length of the road section is 7.5km, the number of bidirectional 6 lanes is 6, and the weather condition and the road condition are good. The road speed limit is 90km/h, the traffic flow is set as a stable flow (15 pcu/km · ln) and is distributed uniformly, and the initial vehicle speed is set as 80km/h. When the experiment is started, the self-vehicle is in an automatic driving mode and runs in a middle lane, the taking-over request time is set to be 5s, and the taking-over request mode adopts a visual and auditory combined mode. 15 drivers were recruited as subjects, and takeover data of 15 groups of the tested drivers were obtained in total.
And filtering the acquired speed and acceleration data, wherein the speed and acceleration tracks after filtering are smoother and more uniform than the actual tracks before filtering, and the fitting degree of the speed and acceleration tracks with the actual tracks is higher.
Based on the data, all the coefficients to be calibrated in the model are calibrated, and the calibration results are shown in table 1.
TABLE 1 calibration result of undetermined constant coefficient in connection pipe risk evaluation model
Calculating the TD and RISK values and the mean value of 15 testees in the taking-over process, carrying out normalization processing on the mean values, and respectively calculating the root mean square error of the 15 testees to obtain that the root mean square error mean value of the RISK is 0.059 and the root mean square error mean value of the TD is 0.093, which shows that the RISK data is superior to the TD data in the aspect of representing the accuracy of RISKs, the error of the RISK data with the actual RISKs is smaller, the data fluctuation is more stable, and the reliability is higher. Therefore, the takeover risk assessment model has reliability in risk assessment, and can effectively assess the takeover risk of the driver.
The above-mentioned disclosure is only a preferred embodiment of the present invention, and is not intended to limit the embodiments of the present invention, and those skilled in the art can easily make various changes and modifications according to the main concept and spirit of the present invention, so the protection scope of the present invention shall be subject to the protection scope of the claims.
Claims (10)
1. A construction method of an automatic driving takeover risk assessment model considering a driving risk field is characterized by comprising the following steps:
step one, calculating a vehicle performance probability factor according to the response speed and the response degree of a vehicle to risks;
step two, constructing a static risk distribution function based on distance;
step three, constructing a dynamic risk distribution function based on the motion state;
step four, considering the influence of the reaction time of the connecting pipe, and calculating the field intensity of the driving risk field; and extracting data information from the vehicle and the interactive vehicle;
fifthly, preprocessing the speed and acceleration data of the vehicle based on a Kalman filtering method;
step six, calibrating the equivalent quality of the takeover risk assessment model; calibrating each to-be-calibrated coefficient of the connecting pipe risk evaluation model; and substituting the calibrated equivalent mass and various to-be-fixed constant coefficients into the automatic driving takeover risk assessment model to obtain the automatic driving takeover risk assessment model based on the driving risk field.
2. The method for constructing the automatic driving takeover risk assessment model considering the driving risk field according to claim 1, wherein in the first step, the calculation formula of the vehicle performance probability factor is as follows:
where ζ represents a coefficient of state of the vehicle, ζ =1 for a normal vehicle, and 0 for a faulty vehicle<ζ<1; c-the magnitude (N) of the depression of the brake pedal of the vehicle in the take-over process; c max -maximum value of the depression amplitude (N) of the brake pedal of the vehicle during take-over; z-the steering wheel angle (°) of the vehicle during take-over; z max -a maximum steering wheel angle (°) of the vehicle during take over; alpha (alpha) ("alpha") 1 ，α 2 To be constantCoefficient, α 1 >0，α 2 >0。
3. The method for constructing the automatic driving takeover risk assessment model considering the driving risk field according to claim 1, wherein in the second step, the distance-based static risk distribution function calculation formula is as follows:
in the formula, M i -equivalent mass (kg) of the vehicle i; kappa-undetermined constant coefficient, kappa>0；d ji -the distance (m) between a point j and the host vehicle i in the road plane; gamma, the descent rate factor, can take the value of 1.
4. The method for constructing the risk assessment model taking over the automatic driving under consideration of the driving risk field according to claim 1, wherein in the third step, the dynamic risk distribution function calculation formula based on the motion state is as follows:
in the formula, M i -equivalent mass (kg) of the vehicle i; a-resultant deceleration (m/s) of the own vehicle i during take-over 2 ) (ii) a Delta-the resultant deceleration reduction factor, delta>0, can take the value of 0.15; theta ij -the angle theta between the connecting line direction of the bicycle i and other traffic participants j in the driving risk field and the positive direction of the x axis in the pipe taking process ij ∈[0°,180°]。
5. The method for constructing the automatic driving takeover risk assessment model considering the driving risk field according to claim 1, wherein in the fourth step, the calculation formula of the field strength of the driving risk field is as follows:
in the formula: s. the i The driving risk field intensity of the own vehicle i;
-a dynamic risk distribution function generated by the traffic unit j at the host vehicle i during take-over;
-a static risk distribution function generated by the traffic unit j at the host vehicle i during take-over; p c -a vehicle performance probability factor; t is t i -take-over reaction time(s) of the host vehicle i in the take-over process; lambda [ alpha ] 1 -the dynamic risk distribution function accounts for weight; lambda [ alpha ] 2 -the weight taken up by the static risk distribution function; Φ — undetermined constant coefficient;
the take over risk index is calculated as follows:
in the formula, RISK i -take over risk index of own car i;
-an average value of the driving risk field strength;
substituting expressions (1) to (3) into expression (4) and combining expression (5) to obtain an automatic driving takeover risk assessment model considering a driving risk field as follows:
6. the method for constructing the automatic driving takeover risk assessment model considering the driving risk field according to claim 1, wherein in the fourth step, the real-time coordinate position, the linear acceleration, the driving speed, the steering wheel angle, the brake pedal opening degree and the takeover reaction time of the driver of the vehicle are extracted; data information from and to the interactive vehicle including the type of interactive vehicle, real-time location coordinates, and speed of the vehicle.
7. The method for constructing the automatic driving takeover risk assessment model considering the driving risk field according to claim 1, wherein in the fifth step, the specific algorithm flow is as follows:
(1) Calculating a state prediction value of a Kalman filtering algorithm:
in the formula (I), the compound is shown in the specification,
-observed variables of the speed and acceleration of the vehicle,
v r -speed of vehicle filtering (m/s); a is a r -vehicle filtered acceleration (m/s) -2 ) (ii) a E-coefficient matrix of system prediction state equation; f-input control item matrix; g-the control quantity of the current state; kappa r -a priori predicting process noise in the process; xi r-1 Time rate of change of acceleration (m/s) -3 )；
The coefficient matrix E of the system prediction state equation, the input control term matrix F and the calculation formula of the control quantity G of the current state are shown as the formula (8):
where T is the inverse(s) of the data sampling frequency.
(2) Calculating a filter gain matrix
Assumption xi r-1 And kappa r All satisfy normal distribution, and the covariance is J and C, respectively, then the filter gain matrix is formula (9):
in the formula (I), the compound is shown in the specification,
-covariance matrix of a priori estimated errors (observations); f-matrix of control items entered; e-coefficient matrix of system prediction state equation; C-Process noise in Prior prediction Process κ r Of the measured data.
(3) The covariance matrix of the a priori estimation errors (observations) is given by equation (3-23):
in the formula, F is the control item matrix of the input;
-the observed variables of the speed and acceleration of the vehicle,
g is the control quantity of the current state; c-time rate of change xi of acceleration r-1 Of the measured data.
(4) The covariance matrix of the a posteriori estimation errors (predicted values) is given by equation (3-23):
wherein, K is an identity matrix; w r -a filter gain matrix; e- (E) -)-a coefficient matrix of the system prediction state equation;
-covariance matrix of a priori estimated errors (observations).
8. The method for constructing the risk assessment model of automatic driving takeover in consideration of the driving risk field according to claim 1, wherein in the sixth step, the equivalent mass is calibrated based on the actual mass, type and speed of the vehicle, and the calculation formula is as follows:
in the formula, m i -actual mass (kg) of the vehicle i; v. of i -the speed of travel (m/s) of the host vehicle i; ρ, u, χ, undetermined constant coefficient.
9. The method for constructing the automatic driving takeover risk assessment model considering the driving risk field according to claim 1, wherein in the sixth step, each undetermined constant coefficient of the takeover risk assessment model is calibrated by using a genetic algorithm, and the genetic parameters are configured as follows: setting the size and scale of a population to be 200; the crossover probability is 0.8; the mutation probability is 0.2, and the iteration times are 300.
10. A model for a method of constructing an automated driving takeover risk assessment model taking into account driving risk situations using any one of claims 1 to 9.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211137773.5A CN115620530A (en) | 2022-09-19 | 2022-09-19 | Automatic driving takeover risk assessment model considering driving risk field and construction method thereof |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211137773.5A CN115620530A (en) | 2022-09-19 | 2022-09-19 | Automatic driving takeover risk assessment model considering driving risk field and construction method thereof |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115620530A true CN115620530A (en) | 2023-01-17 |
Family
ID=84858079
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211137773.5A Pending CN115620530A (en) | 2022-09-19 | 2022-09-19 | Automatic driving takeover risk assessment model considering driving risk field and construction method thereof |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115620530A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116341288A (en) * | 2023-05-25 | 2023-06-27 | 吉林大学 | Heterogeneous traffic epidemic car security field modeling method |
| CN116363905A (en) * | 2023-05-19 | 2023-06-30 | 吉林大学 | Heterogeneous traffic flow converging region lane change time judging and active safety control method |
| CN116453345A (en) * | 2023-06-13 | 2023-07-18 | 吉林大学 | Bus driving safety early warning method and system based on driving risk feedback |
| CN116631221A (en) * | 2023-05-05 | 2023-08-22 | 同济大学 | Monte Carlo simulation-based on-road vehicle running risk quantitative calculation method |
| CN116957345A (en) * | 2023-09-21 | 2023-10-27 | 上海伯镭智能科技有限公司 | Data processing method for unmanned system |
-
2022
- 2022-09-19 CN CN202211137773.5A patent/CN115620530A/en active Pending
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116631221A (en) * | 2023-05-05 | 2023-08-22 | 同济大学 | Monte Carlo simulation-based on-road vehicle running risk quantitative calculation method |
| CN116631221B (en) * | 2023-05-05 | 2024-01-23 | 同济大学 | Monte Carlo simulation-based on-road vehicle running risk quantitative calculation method |
| CN116363905A (en) * | 2023-05-19 | 2023-06-30 | 吉林大学 | Heterogeneous traffic flow converging region lane change time judging and active safety control method |
| CN116363905B (en) * | 2023-05-19 | 2023-09-05 | 吉林大学 | Heterogeneous traffic flow converging region lane change time judging and active safety control method |
| CN116341288A (en) * | 2023-05-25 | 2023-06-27 | 吉林大学 | Heterogeneous traffic epidemic car security field modeling method |
| CN116341288B (en) * | 2023-05-25 | 2023-09-05 | 吉林大学 | Heterogeneous traffic epidemic car security field modeling method |
| CN116453345A (en) * | 2023-06-13 | 2023-07-18 | 吉林大学 | Bus driving safety early warning method and system based on driving risk feedback |
| CN116453345B (en) * | 2023-06-13 | 2023-08-18 | 吉林大学 | Bus driving safety early warning method and system based on driving risk feedback |
| CN116957345A (en) * | 2023-09-21 | 2023-10-27 | 上海伯镭智能科技有限公司 | Data processing method for unmanned system |
| CN116957345B (en) * | 2023-09-21 | 2023-12-08 | 上海伯镭智能科技有限公司 | Data processing method for unmanned system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN115620530A (en) | Automatic driving takeover risk assessment model considering driving risk field and construction method thereof | |
| CN109345020B (en) | Non-signalized intersection vehicle driving behavior prediction method under complete information | |
| CN110488802B (en) | Decision-making method for dynamic behaviors of automatic driving vehicle in internet environment | |
| US11648946B2 (en) | Intelligent vehicle platoon lane change performance evaluation method | |
| CN109885040B (en) | Vehicle driving control right distribution system in man-machine driving | |
| US20190155291A1 (en) | Methods and systems for automated driving system simulation, validation, and implementation | |
| CN103921719B (en) | The interactive commercial vehicle rollover method for early warning of chaufeur and system | |
| CN109910879B (en) | Vehicle safety anti-collision control method combining safe distance and collision time | |
| CN116118780A (en) | Vehicle obstacle avoidance track planning method, system, vehicle and storage medium | |
| CN113548050A (en) | Vehicle running control method, device, system and storage medium | |
| CN114407898B (en) | Road changing path planning method and device, intelligent driving automobile and storage medium | |
| CN113592221B (en) | Road section risk dynamic assessment method based on safety substitution evaluation index | |
| CN114987539A (en) | Individual collision grading early warning method and system for automatic driving automobile based on risk field model | |
| CN109387374A (en) | A kind of lane holding level evaluation method | |
| CN107878453B (en) | A kind of automobile emergency collision avoidance integral type control method for hiding dynamic barrier | |
| Hosseini et al. | Adaptive forward collision warning algorithm for automotive applications | |
| CN114889589A (en) | Intelligent automobile steering and braking cooperative collision avoidance control system and method | |
| CN115009274A (en) | Intelligent networking automobile risk assessment method and personalized decision-making method | |
| CN111275986B (en) | Risk decision device and method for vehicle to autonomously merge into main road in acceleration lane | |
| CN114996116A (en) | Anthropomorphic evaluation method for automatic driving system | |
| Wu et al. | Risk Assessment and Enhancement Suggestions for Automated Driving Systems through Examining Testing Collision and Disengagement Reports | |
| CN112417689A (en) | Man-machine driving-sharing risk control modeling method and driving assisting method | |
| CN113823118A (en) | Intelligent network vehicle lane changing method combining urgency degree and game theory | |
| CN110654458B (en) | Vehicle steering wheel control method and device, computer equipment and vehicle | |
| WO2021109165A1 (en) | Prediction method in automatic driving for performing lane change to overtake leading vehicle in target lane |
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 |