CN113460074B - Automatic driving manual takeover request time adjustment method and system - Google Patents
Automatic driving manual takeover request time adjustment method and system Download PDFInfo
- Publication number
- CN113460074B CN113460074B CN202010238170.9A CN202010238170A CN113460074B CN 113460074 B CN113460074 B CN 113460074B CN 202010238170 A CN202010238170 A CN 202010238170A CN 113460074 B CN113460074 B CN 113460074B
- Authority
- CN
- China
- Prior art keywords
- driver
- takeover
- database
- take
- over
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 42
- 230000009471 action Effects 0.000 claims abstract description 40
- 238000013441 quality evaluation Methods 0.000 claims abstract description 7
- 230000006870 function Effects 0.000 claims description 11
- 238000002474 experimental method Methods 0.000 claims description 10
- 230000008859 change Effects 0.000 claims description 7
- 238000012417 linear regression Methods 0.000 claims description 6
- 238000012544 monitoring process Methods 0.000 claims description 6
- 230000001815 facial effect Effects 0.000 claims description 5
- 238000010606 normalization Methods 0.000 claims description 5
- 230000008569 process Effects 0.000 claims description 5
- 230000007704 transition Effects 0.000 claims description 4
- 229960001948 caffeine Drugs 0.000 claims 1
- RYYVLZVUVIJVGH-UHFFFAOYSA-N trimethylxanthine Natural products CN1C(=O)N(C)C(=O)C2=C1N=CN2C RYYVLZVUVIJVGH-UHFFFAOYSA-N 0.000 claims 1
- 238000004364 calculation method Methods 0.000 description 9
- 238000010586 diagram Methods 0.000 description 9
- 238000001514 detection method Methods 0.000 description 4
- 238000004088 simulation Methods 0.000 description 4
- 238000010276 construction Methods 0.000 description 3
- 238000010219 correlation analysis Methods 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 238000004458 analytical method Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000001303 quality assessment method Methods 0.000 description 2
- 230000035484 reaction time Effects 0.000 description 2
- 230000001133 acceleration Effects 0.000 description 1
- 230000006399 behavior Effects 0.000 description 1
- 230000019771 cognition Effects 0.000 description 1
- 239000000470 constituent Substances 0.000 description 1
- 238000013400 design of experiment Methods 0.000 description 1
- 210000001503 joint Anatomy 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000003032 molecular docking Methods 0.000 description 1
- 230000002028 premature Effects 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 230000001105 regulatory effect Effects 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
- B60W60/00—Drive control systems specially adapted for autonomous road vehicles
- B60W60/005—Handover processes
- B60W60/0057—Estimation of the time available or required for the handover
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
- B60W60/00—Drive control systems specially adapted for autonomous road vehicles
- B60W60/005—Handover processes
- B60W60/0053—Handover processes from vehicle to occupant
Landscapes
- Engineering & Computer Science (AREA)
- Automation & Control Theory (AREA)
- Human Computer Interaction (AREA)
- Transportation (AREA)
- Mechanical Engineering (AREA)
- Traffic Control Systems (AREA)
Abstract
The invention relates to a method and a system for adjusting the time of an automatic driving manual takeover request, which can realize high user experience and takeover quality. The automatic driving manual takeover request timing adjustment method sends a takeover request to a driver before a driving state reaches an automatic driving system performance boundary for switching from an automatic driving mode to a manual driving mode, and comprises the following steps: a step of constructing an individual driver database, in which the driver is recorded as a take-over event after completing taking over once and the data sets corresponding to the take-over events are stored in the individual driver database of the driver; a step of taking over the request prompt, which is to calculate the readiness R of the driver for taking over the driving control right and set the advance time T for sending out the taking over request to the driver; a takeover quality evaluation step of calculating takeover quality P of the takeover event according to actual operation data of a driver; and an individual driver database updating step of updating the action coefficients α ', β' in the individual driver database.
Description
Technical Field
The invention relates to the field of automatic driving technology and ergonomic engineering, in particular to a method and a system for adjusting the time of manual take-over request of automatic driving.
Background
According to the Society of Automotive Engineers (SAE), autopilot technology is classified into six classes (L0-L5), and an L3 class autopilot system can perform certain driving tasks and in some cases monitor the driving environment, but the driver must be ready to regain driving control when the autopilot system makes a request.
Namely, under the L3 level automatic driving condition, when the system reaches a performance boundary (non-emergency condition), such as the conditions of a highway construction area and the like, a manual take-over request needs to be sent out, and the control right of the vehicle is given back to the driver. And the L4 level automatic driving also requires the driver to take over under the working condition that the system cannot handle.
In the taking over process, the sending time of the system taking over request is critical, the taking over prompt time is too early, so that the user experience of a driver can be influenced, the excessive trust (or over-reliance) of the driver can be caused, the attention degree of the surrounding traffic environment is reduced, and further, when the system cannot timely send out the taking over request due to failure, the serious consequence that the driver cannot safely finish taking over can be caused; too late take-over prompt time can lead to poor take-over quality for the driver and even failure to complete take-over tasks. The appropriate take-over time is therefore of great importance for the quality of the manual take-over and for the safety of the vehicle driving.
The existing take-over request timing is usually achieved by setting a fixed advance and issuing a take-over request to the driver at the point when the autopilot system reaches this advance from the system boundary. Studies have shown that driver driving distraction can negatively impact driver driving level and takeover quality.
In the condition that the take-over request advance is a fixed value, the take-over quality is different according to the take-over readiness degree (such as distraction degree) of the driver: under the conditions that a driver has better cognition on traffic environment and the taking over readiness degree is higher, taking over can be safely completed; if the driver is distracted to a high degree and does not fully recognize the traffic environment, the quality of the takeover is reduced, and the takeover cannot be completed safely.
Patent document 1 proposes a control device that monitors a driving state such as a line of sight direction and a driving posture of a driver, and predicts a required take-over time based on a current driving state of the driver.
[ Prior Art literature ]
[ patent literature ]
Patent document 1: japanese patent application No. 6342856
However, in the automatic driving system disclosed in patent document 1, the prediction is made only based on the predicted required takeover time of the current driving state of the driver, which cannot accurately evaluate the required takeover time, resulting in a decrease in the accuracy of the prediction, and as a result, satisfactory user experience and takeover quality of the driver cannot be obtained.
Disclosure of Invention
The invention aims to provide a method and a system for adjusting the time of an automatic driving manual takeover request, which can realize high user experience and takeover quality.
Means for solving the problems
According to the automatic driving manual takeover request timing adjustment method related to the present invention, a takeover request is issued to a driver before reaching an automatic driving system boundary at which a transition from an automatic driving mode to an manual driving mode occurs due to a change in driving environment, thereby reminding the driver that takeover control rights are ready, wherein the method includes:
a step of constructing an individual driver database, in which the driver is recorded as a take-over event after completing take-over once, and data sets corresponding to the take-over events are stored in the individual driver database of the driver, wherein each data set comprises: the takeover quality P of the driver and the takeover readiness degree R of the driver are stored in the individual driver database, and the action coefficient alpha 'related to the takeover quality P and the action coefficient beta' related to the takeover readiness degree R are also stored in the individual driver database;
a takeover request prompting step, namely calculating takeover readiness degree R of a driver for driving control rights according to state data of the driver, and setting advance time T for sending the takeover request to the driver according to the takeover readiness degree R and the action coefficients alpha ', beta' stored in the individual driver database;
a takeover quality evaluation step, namely after a driver finishes takeover driving control right corresponding to the takeover request, calculating takeover quality P of the takeover event according to actual operation data of the driver in the takeover event; and an individual driver database updating step of updating the action coefficients α ', β' in the individual driver database according to the takeover quality P and the takeover readiness degree R of the driver recorded in the data set of the individual driver database.
According to the automatic driving manual takeover request timing adjustment system related to the present invention, a takeover request is issued to a driver before reaching an automatic driving system boundary at which a transition from an automatic driving mode to an manual driving mode occurs due to a change in driving environment, thereby reminding the driver that takeover control is ready, wherein the system includes:
the storage module stores an individual driver database corresponding to each driver, wherein the individual driver database stores a data set corresponding to each take over event of the driver, and each data set comprises: the takeover quality P of the driver and the takeover readiness degree R of the driver are stored in the individual driver database, and the action coefficient alpha 'related to the takeover quality P and the action coefficient beta' related to the takeover readiness degree R are also stored in the individual driver database;
the takeover request prompting module calculates takeover readiness degree R of the driver for driving control rights according to the state data of the driver, and sets advance time T for sending the takeover request to the driver according to the takeover readiness degree R and the action coefficients alpha ', beta' stored in the individual driver database;
the takeover quality evaluation module is used for calculating takeover quality P of the takeover event according to actual operation data of the driver in the takeover event after the driver finishes takeover driving control right corresponding to the takeover request; and
and the database updating module is used for executing the step of updating the action coefficients alpha ', beta' in the individual driver database according to the takeover quality P and the takeover readiness degree R of the drivers recorded in the data set of the individual driver database.
According to the method and the system for regulating the automatic driving manual takeover request opportunity, which are related by the invention, when the takeover readiness degree R of each takeover event and the advance time T for issuing the takeover request are calculated, the action coefficient alpha 'related to the takeover quality P and the action coefficient beta' related to the takeover readiness degree R are added, an individual driver database is constructed to store the historical driving data of the individual driver, namely the takeover quality P of each takeover event and the takeover readiness degree R of the driver, and the action coefficients alpha ', beta' are updated according to the historical data. Through such a step, the action coefficients α ', β' reflecting the operation habits of the individual drivers can be updated in time by learning the historical driving data of the individual drivers, so that a value conforming to the driving habits of each driver can be obtained when calculating the advance time T for issuing the takeover request. A high user experience and take over quality can be achieved.
According to another automatic driving manual takeover request timing adjustment method related to the present invention, a takeover request is issued to a driver before reaching an automatic driving system performance boundary for switching from an automatic driving mode to a manual driving mode due to a driving environment change, thereby reminding the driver that takeover control is ready, wherein the method includes:
a driver state acquisition step of acquiring driving state data of a driver;
and a take-over request prompting step, namely setting a time window length TW according to a driving take-over scene, and calculating the take-over readiness degree R of the driver for the driving control right according to the driving state data acquired by the driver state acquisition step in the time window length of the current moment Tc.
According to another automatic driving manual takeover request timing adjustment system related to the present invention, a takeover request is issued to a driver before reaching an automatic driving system performance boundary for switching from an automatic driving mode to an manual driving mode due to a driving environment change, thereby reminding the driver that the driver is ready to take over driving control, wherein the system includes:
the monitoring module is used for collecting driving state data of a driver;
and the takeover request prompting module is used for setting the time window length TW according to the driving takeover scene, and calculating the takeover readiness degree R of the driver for the driving control right according to the driving state data acquired by the monitoring module in the time window length of the current moment Tc.
In the method and system for adjusting the automatic driving manual takeover request time according to the invention, the time window length T is set according to the driving takeover scene W And the takeover readiness degree R is calculated according to the driver state data acquired within the time window length of the current moment Tc, compared with the situation that the takeover readiness degree R is calculated according to the driving state data of the current moment, the driving state of the driver can be more comprehensively and accurately judged according to the data of a period of time before the takeover event, so that high user experience and takeover quality are realized. In addition, by properly setting the time window length T W The early driver state data before the take over event can be eliminated, so that the interference generated by the data with low relevance to the take over event can be reduced when the readiness degree R of take over is calculated. The user experience and quality of takeover can be further improved.
Drawings
Fig. 1 is a flowchart showing a control method of an automatic driving system according to the present invention.
Fig. 2 is a schematic diagram showing constituent modules of the automatic driving system to which the present invention relates.
Fig. 3 is a schematic diagram showing a monitoring module.
Fig. 4 is a schematic diagram showing a calculation and adjustment module.
FIG. 5 is a schematic diagram illustrating a take over request hint module.
Fig. 6 is a schematic diagram illustrating a take over quality assessment module.
Fig. 7 is a schematic diagram showing an embodiment takeover scenario.
Fig. 8 is a diagram showing a result of a correlation analysis of the driver's takeover readiness and the takeover quality.
Fig. 9 is a diagram showing a result of a correlation analysis of the driver's takeover readiness and the subjective score of the degree of driving distraction.
Fig. 10 is a diagram showing a correlation analysis result of the take-over request issue advance time and the take-over quality.
Detailed Description
The first embodiment of the invention is a method for adjusting the request opportunity of automatic driving manual takeover, and the whole flow is shown in fig. 1. The method comprises the following steps.
Step S1: step of constructing driver big data cloud database
Constructing a driver big data cloud database through a driving simulator experiment and a real vehicle experiment based on a taking over scene, wherein parameters stored in the database comprise P 0 、α、β、T 0 、R 0 Each parameter satisfies the formula: p (P) 0 ＝αT 0 +βR 0 。
Wherein P is 0 The method is based on the target takeover quality of experimental big data, and can ensure the takeover quality of the safe takeover of the driver, and can take 6 to 12 seconds based on experimental data and different takeover scenes; alpha is the action coefficient of taking over request sending advance time to take over quality of driver obtained by driving simulation experiment, beta is the action coefficient of taking over readiness degree of driver obtained by driving simulation experiment to the quality of the butt joint pipe, and alpha and beta are obtained by performing multiple linear regression analysis on the data pair obtained by driving simulation experiment; t (T) 0 The initial take-over request sends out the advance time, namely the minimum take-over request advance time which can be safely taken over based on driving simulation big data statistics, namely the take-over advance time required when a driver fully focuses on observing the traffic environment; r is R 0 The driver is completely focused on the readiness of the takeover when observing the traffic environment, and the value is 100%.
Step S2: step of constructing an individual driver database
Constructing an individual driver database by using real vehicle takeover data of driven vehicles (hereinafter referred to as 'own vehicles'), wherein the individual driver databases of different vehicles are obviously different, and the completion of takeover by the driver is recorded as a takeover event, and the parameters stored in the database comprise alpha ', beta', T i 、T 0 Data set { P 'corresponding to each take over event' 0 ，ΔT，T p ，P，R}。
Alpha' is when the takeover request of the host vehicle is sent out in advanceTaking alpha and beta' as initial values of the action coefficients of the connecting pipe quality of the driver, wherein the initial values of the action coefficients are the action coefficients of the connecting pipe quality of the connecting pipe readiness degree of the driver of the vehicle, and taking beta as the initial values of the action coefficients; t (T) i Is a takeover time adjustment item of an individual driver, and the initial value of the takeover time adjustment item is selected to be 5-10 s for ensuring safety. Data set { P 'corresponding to each take over event' 0 ，ΔT，T p In P, R }, P' 0 The target taking quality is that the initial value is P in the cloud database of the big data of the driver 0 Delta T is the error in take over hint time, T p The posterior takeover advance time is that R is the driver's readiness to take over, and P is the actual takeover mass in seconds.
Step S3: step of judging performance boundary of automatic driving system
Judging whether the performance boundary of the automatic driving system exists in the automatic driving (namely, the working condition which cannot be processed when the automatic driving system of the L3 and L4 levels is in the automatic driving mode). If so, go to step 4. If not, the method is exited.
Step S4: driver state acquisition step
According to the difference of detection means, the state of the driver is mainly classified into two types, one type is bioelectric signals measured by a contact type device, such as electrocardio, electroencephalogram, etc., and the other type is information measured by a non-contact type device, such as face information, voice information, etc., of the driver.
According to the invention, through non-contact measurement, the pitching angle and the yaw angle of the face of the driver are selected as the states of the driver, the face orientation detection equipment collects the yaw angle pitch and the pitching angle yaw of the face of the driver at the same time at a set frequency, the pitch and the yaw collected at the same time are taken as one face data point, and the yaw angle and the pitching angle of the face of the driver during manual driving are taken as references.
Step S5: take over request prompting step
First, the readiness R of the driver for takeover is calculated. By a through-normalization function S d Normalizing the actually collected facial data points if the face orientation of the driver is within the manual driving operation rangeIf the driver is not distracted, otherwise the driver is distracted, the function S is normalized d The expression of (2) is as follows:
taking into account the timeliness of the degree of driver distraction, i.e. premature distraction does not affect the current driving behaviour, the weighting function W is passed through a time window t Weighting the takeover readiness level R, takeover readiness levels R and W t The expression of (2) is as follows:
wherein f t The frequency of data acquisition for the face orientation detection device, i.e. the frequency of data acquisition for the face orientation detection device; t (T) W For the time window length, according to different takeover scene settings, the more complex the takeover scene is, the time window length T W Longer, preferably T W 8s-15s; t (T) c For the current time, t is the time corresponding to each facial data point acquired in the current driving process, ΣS d Representing a normalization function S corresponding to all facial data points acquired within the length of the time window d Summing is performed.
Then, calculating the current takeover request advance time T according to the current takeover readiness degree R of the driver, wherein the calculation formula is as follows:
the distance to boundary time TTB is calculated from the vehicle speed and the distance from the driving system boundary (step S5A).
And judging whether to send a takeover request to a driver according to the time TTB required by the vehicle to run to the performance boundary of the automatic driving system (namely, the working condition which cannot be processed when the L3 and L4-level automatic driving system is in the automatic driving mode) and the current takeover request advance time T (step S5B).
Comparing T with TTB, if T is more than or equal to TTB, sending a take-over request to a driver, and executing step S6; if T is smaller than TTB, continuing waiting until T is larger than or equal to TTB, sending a take-over request to a driver, and executing step S6.
Step S6: take over quality assessment step
The driver completes the current takeover according to the current takeover request (step S6A). The current actual takeover mass P is then calculated from the driver' S current takeover operation data (step S6B). The index of the take-over quality is selected from various types, such as a statistical value of input data of a steering wheel of a driver, a statistical value of input data of an accelerator pedal of the driver, a statistical value of input data of a decelerator pedal of the driver, a reaction time of the driver, and the like. The time interval TTBT of the boundary between the self-vehicle and the automatic driving system and the difference between the TOT of the reaction time of the driver are taken as the quality of the take-over, namely:
P＝TTBT-TOT
in particular, when the takeover fails, i.e., a collision or the like occurs, P takes 0. Meanwhile, the posterior takeover advance time T is calculated by p ：
I.e. T p Is the take-over advance time calculated from the actual take-over quality.
Step S7: recording take-over time data set
After the current take-over event occurs, calculating a take-over prompt time error delta T corresponding to the current take-over event, wherein delta T=T p T, the take-over cue time error Δt, is the difference between the a-priori take-over advance time and the take-over request advance time. Since T is used for ensuring safety i Is selected from 5s to 10s. Recording data { P 'corresponding to the current take-over event' 0 ，ΔT，T p ，P，R}。
In addition, the data recorded in the individual driver database is uploaded to the driver big data cloud database, and the parameters P 'stored in the driver big data cloud database as the individual driver database are updated periodically according to the uploaded data' 0 、α′、β′、T i 、T 0 Parameter P of initial value 0 、α、β、T 0 、R 0 。
Step S8: individual driver database updating step
Judging { P 'in the current individual driver database' 0 ，ΔT，T p Whether the total group number N of P, R reaches an integer multiple of N1 (step S8A). If yes, updating T in the current individual driver database i ，T i Taking the statistical values of all delta T in the current individual driver database, such as the mean value or the median of delta T, and then entering the following step 8B; otherwise, not updating T in the current individual driver database i The process returns to step S2.
If the value of N1 is too large, the data update is not timely, and the driver is excessively trusted (because the value of Δt is generally negative to ensure safety); if the value of N1 is too small, the data will fluctuate greatly and be unstable, so based on the data obtained by the experiment, N1 should take a value in the range of 8-12, preferably 10.
9) Judging { P 'in the current individual driver database' 0 ，ΔT，T p Whether the total group number N of P, R reaches an integer multiple of N2 (step S8B). If yes, a multiple linear regression method is adopted, and alpha 'and beta' in the current individual driver database are updated through the following formulas: p=α' (t+t) i ) +β' R, and then returns to step S2. Otherwise, the alpha 'and beta' in the current individual driver database are not updated, and the step S2 is directly returned.
It should be noted that if the value of N2 is too large, the data update will not be timely, and thus the data will be excessively trusted (because the value of Δt is generally negative to ensure security); if the value of N2 is too small, the data will fluctuate greatly and be unstable, so based on the data obtained by the experiment, N2 should take a value in the range of 450-550, preferably 500.
Further, in step S8A, if { P 'is currently in the individual driver database' 0 ，ΔT，T p When the total group number N of P and R reaches the integral multiple of N1, firstly judging the delta T calculated in the step S6, and removing the data group { P 'corresponding to the delta T within the range of three times standard deviation of normal distribution' 0 ，ΔT，T p P, R, then update T in the current individual driver database i ，T i And (2) taking the statistical values of all delta T in the current individual driver database, such as the mean value, standard deviation, median and the like of the delta T, and returning to the step (S2).
According to the above embodiment, the following advantageous effects can be obtained.
(1) In the automatic driving manual takeover request timing adjustment method and system, when the takeover readiness degree R of each takeover event and the advance time T for issuing the takeover request are calculated, the action coefficient alpha 'related to the takeover quality P and the action coefficient beta' related to the takeover readiness degree R are added, the historical driving data of individual drivers, namely the takeover quality P of each takeover event and the takeover readiness degree R of the drivers, are built in the individual driver database, and the action coefficients alpha ', beta' are updated according to the historical data. Through such a step, the action coefficients α ', β' reflecting the operation habits of the individual drivers can be updated in time by learning the historical driving data of the individual drivers, so that a value conforming to the driving habits of each driver can be obtained when calculating the advance time T for issuing the takeover request. A high user experience and take over quality can be achieved.
(2) Since the data recorded in the individual driver database is uploaded to the driver big data cloud database, the parameter P 'stored in the driver big data cloud database as the individual driver database is updated periodically according to the uploaded data' 0 、α′、β′、T i 、T 0 Parameter P of initial value 0 、α、β、T 0 、R 0 This way more can be done based on more driversAnd taking over the initial value of the data updating cloud database, and providing a better initial value for the vehicles produced in the future.
(3) In the above-mentioned takeover request prompting step, since the time window length T is set according to the driving takeover scenario W And the takeover readiness degree R is calculated according to the driver state data acquired within the time window length of the current moment Tc, compared with the situation that the takeover readiness degree R is calculated according to the driving state data of the current moment, the driving state of the driver can be more comprehensively and accurately judged according to the data of a period of time before the takeover event, so that high user experience and takeover quality are realized. In addition, by properly setting the time window length T W The early driver state data before the take over event can be eliminated, so that the interference generated by the data with low relevance to the take over event can be reduced when the readiness degree R of take over is calculated. The user experience and quality of takeover can be further improved.
(4) In the individual driver database updating step, the coefficient of action α ', β ', p=α ' (t+t) is updated by using a multiple linear regression method and by a formula i ) The +beta' R can simply obtain data which can accurately reflect the driving habit and state of an individual driver, thereby calculating the proper taking over readiness degree R and the advance time T for sending out the taking over request. In addition, the take-over time adjustment term T of the individual driver is introduced into the updated formula i By adjusting this parameter according to the driver, the safety of the takeover can be further ensured.
(5) Docking the time adjustment item T in an individual driver database update step i Updating, in particular the take-over time adjustment term T i Taking the average of all deltat's in the current individual driver database. This makes it possible to effectively use the driving history data of the driver for the management time adjustment item T i Timely updating is performed, so that the driving habit of the driver is better reflected in the calculation of the readiness for take over R and the advance time T for making the take over request.
(6) The total number of groups N in the data group reaches N1 or N2Respectively updating the takeover time adjustment items T when the integral multiple is adopted i And the action coefficients alpha ', beta', and respectively set the N1 and N2 as a numerical value in the range of 8-12 and a numerical value in the range of 450-550, so that the data can be ensured to be updated in time, and the data is prevented from being greatly fluctuated and unstable.
(7) By collecting the face pitch angle yaw and the yaw angle pitch of the driver as the state data of the driver, the degree of distraction of the driver can be accurately detected, thereby more reasonably calculating the readiness degree R.
Other variants
In the above embodiment, the method of multiple linear regression is adopted and the method is performed by the formula p=α' (t+t i ) The +β ' R updates the acting coefficient α ', β ', but the acting coefficient α ', β ' may be updated by other formulas as long as the driving history of the driver can be reflected.
In the above embodiment, the individual driver database also stores the takeover time adjustment item T of the driver i However, the takeover time adjustment item T may not be stored on the premise of ensuring the takeover security i 。
In the above embodiment, the parameter P 'stored in the driver big data cloud database as the individual driver database is updated periodically' 0 、α′、β′、T i 、T 0 Parameter P of initial value 0 、α、β、T 0 、R 0 However, if the individual driver database is used alone, the driving history of the driver can be accurately reflected in the update of the action coefficients α ', β', and the initial values of the respective parameters may be set by other methods without using a large data cloud database.
In the above embodiment, the face pitch angle yaw and the yaw angle pitch of the driver are collected as the state data of the driver, but other parameters of the driver may be used as the state data, such as the positions of the hands and feet of the driver, the driving posture, and the like.
A second embodiment of the invention is an autopilot system. The system mainly comprises a storage module, a monitoring module, a calculating and adjusting module, a take-over request prompting module, a take-over quality evaluation module and a system parameter online learning module, as shown in fig. 2.
1) And a storage module: the control method comprises a driver big data cloud database and an individual driver database, wherein data stored in the two databases are described in the control method, and are not repeated herein.
2) And a monitoring module: the pitching angle and the yaw angle of the face of the driver are acquired by the camera, and the takeover readiness degree (R) of the driver is output in real time after processing and calculation. Mainly comprising the functions of face recognition, face orientation recognition, driver's takeover readiness (R) calculation, as shown in fig. 3.
3) And the calculating and adjusting module is used for: including take over request advance time (T) computation, time to system boundary (TTB) computation, take over request decision function, as shown in fig. 4. The calculation formula is as described above, and is not described here.
4) The take over request prompting module: and sending out a take-over request to a driver in a mode of prompt tone, head-up display/instrument panel graphic prompt and the like according to the take-over request decision of the calculation and adjustment module until the take-over of the driver is completed, as shown in fig. 5.
5) Taking over the quality evaluation module: from the take-over prompt sending time to the take-over completion time, the take-over quality is calculated according to the original information of the vehicle-mounted sensor information, such as acceleration, angular velocity, steering wheel angle, pedal travel and the like, as shown in fig. 6.
6) And the system parameter online learning module is used for: and updating the system parameter values of the individual driver database in the storage module according to the rules by the obtained takeover quality result and the takeover readiness result.
The present embodiment can obtain advantageous effects corresponding to those of the first embodiment.
The validity of the invention is verified in connection with the following examples:
1) Taking over a scene
And selecting one of typical boundaries of the current L3 level automatic driving system, namely a highway construction area, as a boundary of the takeover. The traffic scene of the connecting pipe is set to be a three-lane expressway, and in the daytime of sunny days, the expressway only leaves one lane to pass (leftmost lane or rightmost lane) due to the construction of two closed lanes. The driver needs to perform a lane change operation after receiving the take-over prompt, and before receiving the take-over request, the vehicle is in an automatic driving mode, and automatically keeps a speed of 100 km/h to drive in the middle lane. Random traffic flow occurs during automatic driving, traffic flow exists in a target lane during lane changing, two vehicles are respectively positioned in front of and behind an own vehicle, and lane changing is random left and right during taking over each time so as to avoid learning effect, as shown in fig. 7.
2) Subjective assessment of driver readiness to take over
The driver needs to score the distraction of this drive (an integer of 0-10), 0 as least distraction, and 10 as most distraction after each take over is completed. And the data of each driver is standardized, namely, firstly, a scoring mean value of a plurality of times of taking over experiments of a certain driver is obtained, and then, the mean value is subtracted from the original data to obtain the final-use scoring of the distraction degree of the driver.
3) Design of experiment
The driving simulator is selected as a test platform, the data of 16 Chinese drivers are tested, and the taking over prompt time is divided into three types: 6 seconds before the boundary, 8 seconds before the boundary, and 10 seconds before the boundary. A total of 18 takeover experiments were performed for each driver in 3 groups. And the driver is visually distracted by playing the video, and the driver decides to watch the video or the surrounding environment by himself.
Collecting yaw angle and pitch angle (frequency of 20 Hz) of face orientation of a driver through a monocular camera in the vehicle, and normalizing the actually collected yaw angle pitch and pitch angle yaw by taking the yaw angle and pitch angle of manual driving of the driver as a reference, namely considering that the face orientation of the driver is not distracted if the face orientation of the driver is in the range of manual driving, or considering that the face orientation of the driver is distracted if the face orientation of the driver is not distracted, and using a normalization function S d And (3) performing standardization:
taking into account the timeliness of the driver distraction level, the weighting function W is passed through again t Weighting, time window length W t Take 12 seconds:
i.e. the yaw angle and pitch angle of the original face orientation of the driver are normalized (S d ) And a time window weighting function (W t ) The driver takes over the readiness level after weighting is obtained as shown in fig. 8. Wherein, the calculation formula of the take-over degree R is as follows:
the judging conditions of the taking over time are as follows: the absolute value of the steering wheel angle is larger than 2 degrees, the travel of the accelerator pedal is larger than 5%, and the travel of the decelerator pedal is larger than 5%. When any one of the judging conditions is reached, the system judges that the take-over event occurs.
4) Experimental results and analysis
After the original data are processed, the takeover quality P of the driver is obtained through calculation, and the readiness degree R of the driver takeover and the distraction degree subjective scoring of the driver are obtained.
4.1 driver takes over readiness and driver subjective distraction scoring
As shown in fig. 9, the driver take over level is inversely related to the driver's subjective distraction score, significance level p<0.001, determination coefficient r 2 =0.436, it is reasonable and effective to explain the method of evaluating the driver's readiness for takeover based on visual distraction proposed in the present invention.
4.2 driver takeover readiness and driver takeover quality (proof that R has an impact on P)
Based on experimental data, analyzing the correlation between the readiness degree of the takeover of the driver and the takeover quality P of the driver, and obtaining a result as a significance levelp<0.001, determination coefficient r 2 =0.309 as shown in fig. 8, wherein,
P＝TTBT-TOT
the results indicate that the driver's readiness to take over is related to the quality of the driver's take over.
4.3 take over request Advance time and driver take over quality (proof T has an impact on P)
Based on experimental data, the take-over request advance time (T) and the driver take-over mass (P) are analyzed by variance. Data at the same takeover readiness level were selected from the experimental results and examined for takeover quality at different takeover request issue advance times, resulting in a significance level p <0.001, f value= 41.535, as shown in fig. 10. It is illustrated that the regulation of the quality of the butt-joint pipe by taking over the advance time of the request issuing is effective. In the case of takeover scenario determination, the takeover quality of the driver is related to the readiness of the takeover of the driver and the advance time of the takeover request, and the more advanced the takeover prompt, the more advantageous is for high quality takeover completion.
In combination with the foregoing, when the readiness of the takeover of the driver is low, the takeover quality can be improved by increasing the advance time of the takeover request. Further, different take-over request advance times may be provided according to the driver take-over readiness, and thus similar or identical target take-over quality can be achieved by adjusting the take-over request advance time in any driver take-over readiness.
Claims (22)
1. An automatic driving manual takeover request timing adjustment method that issues a takeover request to a driver before a driving state reaches an automatic driving system performance boundary at which a transition from an automatic driving mode to a manual driving mode occurs due to a driving environment change, thereby reminding the driver that takeover control is ready, comprising:
a step of constructing an individual driver database, in which the driver is recorded as a take-over event after completing take-over once, and data sets corresponding to the take-over events are stored in the individual driver database of the driver, wherein each data set comprises: the takeover quality P of the driver and the takeover readiness degree R of the driver are stored in the individual driver database, and the action coefficient alpha 'related to the takeover quality P and the action coefficient beta' related to the takeover readiness degree R are also stored in the individual driver database;
a takeover request prompting step, namely calculating takeover readiness degree R of a driver for driving control rights according to state data of the driver, and setting advance time T for sending the takeover request to the driver according to the takeover readiness degree R and the action coefficients alpha ', beta' stored in the individual driver database;
a takeover quality evaluation step, namely after a driver finishes takeover corresponding to the takeover request, calculating takeover quality P of the takeover event according to actual operation data of the driver in the takeover event; and
an individual driver database updating step of updating the action coefficients α ', β' in the individual driver database according to the takeover quality P and the takeover readiness R of the driver stored in the data set of the individual driver database.
2. The automatic driving manual takeover request timing adjustment method according to claim 1, wherein,
the method further comprises the step of constructing a driver big data cloud database, wherein the driver big data cloud database is constructed through a driving simulator experiment and a real vehicle experiment based on a take-over scene, and the take-over quality P, the action coefficients alpha ', beta', the advance time T of a take-over request and the initial value of the take-over readiness degree R are obtained from parameters P stored in the driver big data cloud data 0 、α、β、T 0 、R 0 Each parameter satisfies the formula:
P 0 ＝αT 0 +βR 0 。
3. the automatic driving manual takeover request timing adjustment method according to claim 2, wherein,
in constructing theIn the step of constructing the driver big data cloud database, the data of all the individual driver databases constructed in the step of constructing the individual driver databases are uploaded to the driver big data cloud database, so that the parameters P stored in the driver big data cloud database are updated 0 、α、β、T 0 、R 0 。
4. The automatic driving manual takeover request timing adjustment method according to claim 1, wherein,
the individual driver database also stores a takeover time adjustment item T of the driver i In the individual driver database updating step, the coefficient of action alpha ', beta' is updated by using a multiple linear regression method and by the following formula,
P＝α′(T+T i )+β′R。
5. the automatic driving manual takeover request timing adjustment method according to claim 4, wherein,
each of said data sets further comprises a target take over mass P' 0 In the take-over request prompting step, the take-over request advance time T is calculated according to the following formula,
6. the automatic driving manual takeover request timing adjustment method according to claim 5, wherein,
each of the data sets further includes a posterior nozzle advance time T p And take over the hint time error Δt, which satisfies:
ΔT＝T p -T。
7. the automatic driving manual takeover request timing adjustment method according to claim 6, wherein,
in the individualThe step of updating the body driver database further comprises updating the takeover time adjustment item T i In which the take-over time adjustment term T is applied i Taking the average of all deltat's in the current individual driver database.
8. The automatic driving manual takeover request timing adjustment method according to claim 7, wherein,
updating the take-over time adjustment term T when the total number of groups n of the data groups in the individual driver database reaches an integer multiple of a first prescribed value i Is carried out by a method comprising the steps of.
9. The automatic driving manual takeover request timing adjustment method according to claim 1, wherein,
updating the coefficient of action α ', β' in the individual driver database when the total number of groups n of the data groups in the individual driver database reaches an integer multiple of a second prescribed value.
10. The automatic driving manual takeover request timing adjustment method according to claim 1, wherein,
further comprises a driver state acquisition step of acquiring a yaw angle pitch and a pitch angle yaw of the face of the driver as state data of the driver,
in the takeover request prompting step, the takeover readiness degree R is calculated using the driver state data acquired in the driver state acquisition step.
11. The method for adjusting the timing of an automatic pilot manual takeover request according to claim 10,
the take over readiness is calculated according to the following formula,
wherein f t The acquisition frequency of the driver state data is the frequency of acquiring yaw angle pitch and pitch angle yaw data of the face of the driver;
T W is the time window length;
T c is the current moment;
t is the time corresponding to each face data point collected in the current driving process;
S d representing a value normalized by the actual collected facial data points, a normalization function S d The expression of (2) is as follows:
12. an automated manual takeover request timing adjustment system that issues a takeover request to a driver before reaching an automated driving system performance boundary that transitions from an automated driving mode to a manual driving mode due to a driving environment change, thereby alerting the driver that takeover control is ready, comprising:
the storage module stores an individual driver database corresponding to each driver, the individual driver database stores a data set corresponding to each take over event of the driver, and each data set comprises: the takeover quality P of the driver and the takeover readiness R of the driver are stored in the individual driver database, and the action coefficient alpha 'related to the takeover quality P and the action coefficient beta' related to the takeover readiness R are also stored in the individual driver database;
the takeover request prompting module calculates takeover readiness degree R of the driver for driving control rights according to the state data of the driver, and sets advance time T for sending the takeover request to the driver according to the takeover readiness degree R and the action coefficients alpha ', beta' stored in the individual driver database;
the takeover quality evaluation module is used for calculating takeover quality P of the takeover event according to actual operation data of the driver in the takeover event after the driver finishes takeover driving control right corresponding to the takeover request; and
and the database updating module is used for updating the action coefficients alpha ', beta' in the individual driver database according to the takeover quality P and the takeover readiness degree R of the drivers stored in the data set of the individual driver database.
13. The automated driving manual take over request timing adjustment system according to claim 12, wherein,
the storage module is further stored with a driver big data cloud database which is constructed through a driving simulator experiment and a real vehicle experiment based on a take-over scene, and the take-over quality P, the action coefficients alpha ', beta', the advance time T of the take-over request and the initial value of the take-over readiness degree R are obtained and stored in the parameter P stored in the driver big data cloud data 0 、α、β、T 0 、R 0 Each parameter satisfies the formula:
P 0 ＝αT 0 +βR 0 。
14. the automated driving manual take over request timing adjustment system according to claim 13, wherein,
uploading the data of all the individual driver databases to the driver big data cloud database so as to update the parameter P stored in the driver big data cloud database 0 、α、β、T 0 、R 0 。
15. The automated driving manual take over request timing adjustment system according to claim 12, wherein,
the individual driver databaseAlso stored is a take-over time adjustment item T for the driver i The database updating module adopts a multiple linear regression method and updates the action coefficients alpha ', beta',
P＝α′(T+T i )+β′R。
16. the automated driving manual take over request timing adjustment system of claim 15, wherein,
each of said data sets further comprises a target take over mass P' 0 The take-over request prompting module calculates the take-over request advance time T through the following formula:
17. the automated driving manual take over request timing adjustment system of claim 16, wherein,
each of the data sets further includes a posterior nozzle advance time T p And take over the hint time error Δt, which satisfies:
ΔT＝T p -T。
18. the automated driving manual take over request timing adjustment system of claim 17, wherein,
the database updating module further adjusts the T of the takeover time i Updating to make the takeover time adjustment item T i The average of all deltat's in the individual driver database is taken.
19. The automated driving manual take over request timing adjustment system of claim 18, wherein,
when the total group number n of the data groups in the individual driver database reaches an integer multiple of a first prescribed value, the database updating module updates the takeover time adjustment item T i 。
20. The automated driving manual take over request timing adjustment system according to claim 12, wherein,
the database update module updates the action coefficients α ', β' in the individual driver database when the total number of groups n of the data groups in the individual driver database reaches an integer multiple of a second prescribed value.
21. The automated driving manual take over request timing adjustment system according to claim 12, wherein,
further comprises a monitoring module for collecting the yaw angle pitch and the pitch angle yaw of the face of the driver as state data of the driver,
the takeover request prompting module calculates the takeover readiness level R using the driver state data acquired in the driver state acquisition step.
22. The automated driving manual take over request timing adjustment system of claim 21, wherein,
the take over request prompting module calculates the take over readiness level R according to the following formula,
wherein f t The acquisition frequency of the driver state data is the frequency of acquiring yaw angle pitch and pitch angle yaw data of the face of the driver;
T W is the time window length;
T c is the current moment;
t is the time corresponding to each face data point collected in the current driving process;
S d representing a value normalized by the actual collected facial data points, a normalization function S d The expression of (2) is as follows:
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010238170.9A CN113460074B (en) | 2020-03-30 | 2020-03-30 | Automatic driving manual takeover request time adjustment method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010238170.9A CN113460074B (en) | 2020-03-30 | 2020-03-30 | Automatic driving manual takeover request time adjustment method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113460074A CN113460074A (en) | 2021-10-01 |
| CN113460074B true CN113460074B (en) | 2023-11-07 |
Family
ID=77864921
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010238170.9A Active CN113460074B (en) | 2020-03-30 | 2020-03-30 | Automatic driving manual takeover request time adjustment method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113460074B (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105809152A (en) * | 2016-04-06 | 2016-07-27 | 清华大学 | Monitoring method for cognitive distraction of driver on basis of multi-source information fusion |
| CN107662611A (en) * | 2017-11-06 | 2018-02-06 | 吉林大学 | A kind of automatic driving mode switching system based on driver's Emotion identification |
| DE102017215714A1 (en) * | 2017-09-06 | 2019-03-07 | Bayerische Motoren Werke Aktiengesellschaft | Method and system for improving driver adoption |
| CN110626336A (en) * | 2019-09-24 | 2019-12-31 | 中国第一汽车股份有限公司 | Control system for automatic driving of L3-grade vehicle |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2019034576A (en) * | 2017-08-10 | 2019-03-07 | オムロン株式会社 | Driver state recognition apparatus, driver state recognition system, and driver state recognition method |
-
2020
- 2020-03-30 CN CN202010238170.9A patent/CN113460074B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105809152A (en) * | 2016-04-06 | 2016-07-27 | 清华大学 | Monitoring method for cognitive distraction of driver on basis of multi-source information fusion |
| DE102017215714A1 (en) * | 2017-09-06 | 2019-03-07 | Bayerische Motoren Werke Aktiengesellschaft | Method and system for improving driver adoption |
| CN107662611A (en) * | 2017-11-06 | 2018-02-06 | 吉林大学 | A kind of automatic driving mode switching system based on driver's Emotion identification |
| CN110626336A (en) * | 2019-09-24 | 2019-12-31 | 中国第一汽车股份有限公司 | Control system for automatic driving of L3-grade vehicle |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113460074A (en) | 2021-10-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112041910B (en) | Information processing apparatus, mobile device, method, and program | |
| CN109572550B (en) | Driving track prediction method, system, computer equipment and storage medium | |
| US9676395B2 (en) | Incapacitated driving detection and prevention | |
| EP1332910B1 (en) | Method and system for vehicle operator assistance improvement | |
| EP1997705B1 (en) | Drive behavior estimating device, drive supporting device, vehicle evaluating system, driver model making device, and drive behavior judging device | |
| US20090040054A1 (en) | Real-time driving danger level prediction | |
| CN113071505B (en) | Method, device and equipment for determining driving behavior habit and controlling vehicle running | |
| CN110386145A (en) | A kind of real-time forecasting system of target driver driving behavior | |
| Sun et al. | An integrated solution for lane level irregular driving detection on highways | |
| US11040720B2 (en) | Sleepiness level prediction device and sleepiness level prediction method | |
| CN113516846B (en) | Vehicle lane change behavior prediction model construction, prediction and early warning method and system | |
| CN114239423A (en) | Method for constructing prediction model of danger perception capability of driver on long and large continuous longitudinal slope section | |
| Toledo et al. | State dependence in lane-changing models | |
| CN114049677A (en) | Vehicle ADAS control method and system based on emotion index of driver | |
| CN112319505B (en) | Automatic driving lane change judgment method, device, equipment and storage medium | |
| CN112462759B (en) | Evaluation method, system and computer storage medium of rule control algorithm | |
| CN113460074B (en) | Automatic driving manual takeover request time adjustment method and system | |
| CN111401217B (en) | Driver attention detection method, device and equipment | |
| Peng et al. | Lane changing decision based on binary finite and zero-sum grey game | |
| CN112580201B (en) | Simulation test evaluation method and system for vehicle emergency steering auxiliary system | |
| CN115169477A (en) | Method, device and equipment for evaluating safety sense of assistant driving system and storage medium | |
| CN114148349A (en) | Vehicle personalized following control method based on generation countermeasure simulation learning | |
| CN112530164B (en) | Active driving safety early warning method and system for high-altitude area road | |
| CN115027497A (en) | Target vehicle cut-in intention prediction method and readable storage medium | |
| CN113415292A (en) | Driving takeover capability evaluation method and electronic 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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |