CN114633743B - Automatic driving vehicle and collision accident detection method and system thereof - Google Patents
Automatic driving vehicle and collision accident detection method and system thereof Download PDFInfo
- Publication number
- CN114633743B CN114633743B CN202011487872.7A CN202011487872A CN114633743B CN 114633743 B CN114633743 B CN 114633743B CN 202011487872 A CN202011487872 A CN 202011487872A CN 114633743 B CN114633743 B CN 114633743B
- Authority
- CN
- China
- Prior art keywords
- vehicle
- acceleration
- longitudinal acceleration
- deceleration
- collision accident
- 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
- 238000001514 detection method Methods 0.000 title claims abstract description 83
- 230000001133 acceleration Effects 0.000 claims abstract description 277
- 230000004044 response Effects 0.000 claims abstract description 42
- 238000004364 calculation method Methods 0.000 claims abstract description 36
- 238000000034 method Methods 0.000 claims description 48
- 230000008569 process Effects 0.000 claims description 15
- 238000012360 testing method Methods 0.000 claims description 6
- 230000000875 corresponding effect Effects 0.000 description 45
- 230000008859 change Effects 0.000 description 12
- 230000009286 beneficial effect Effects 0.000 description 5
- 230000001934 delay Effects 0.000 description 5
- 230000008901 benefit Effects 0.000 description 4
- 238000007405 data analysis Methods 0.000 description 4
- 206010039203 Road traffic accident Diseases 0.000 description 3
- 238000013459 approach Methods 0.000 description 3
- 230000007423 decrease Effects 0.000 description 3
- 238000005259 measurement Methods 0.000 description 3
- 230000003068 static effect Effects 0.000 description 3
- 238000004422 calculation algorithm Methods 0.000 description 2
- 230000002596 correlated effect Effects 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 239000006185 dispersion Substances 0.000 description 2
- 230000010355 oscillation Effects 0.000 description 2
- 238000005381 potential energy Methods 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 239000000725 suspension Substances 0.000 description 2
- 238000012795 verification Methods 0.000 description 2
- 238000011161 development Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 230000004927 fusion Effects 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 230000000737 periodic effect Effects 0.000 description 1
- 238000004088 simulation Methods 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
- B60W30/00—Purposes of road vehicle drive control systems not related to the control of a particular sub-unit, e.g. of systems using conjoint control of vehicle sub-units
- B60W30/08—Active safety systems predicting or avoiding probable or impending collision or attempting to minimise its consequences
- B60W30/095—Predicting travel path or likelihood of collision
- B60W30/0953—Predicting travel path or likelihood of collision the prediction being responsive to vehicle dynamic parameters
Landscapes
- Engineering & Computer Science (AREA)
- Automation & Control Theory (AREA)
- Transportation (AREA)
- Mechanical Engineering (AREA)
- Traffic Control Systems (AREA)
- Regulating Braking Force (AREA)
- Control Of Driving Devices And Active Controlling Of Vehicle (AREA)
Abstract
The invention provides an automatic driving vehicle and a collision accident detection method and system thereof, and belongs to the technical field of automatic driving. The collision accident detection method comprises the following steps: determining the running state of the current vehicle according to the vehicle speed, the accelerator opening and the brake deceleration at the current moment, wherein the running state of the vehicle comprises at least one of an acceleration state, a deceleration state and an emergency brake stopping state; calculating an expected value of the longitudinal acceleration of the vehicle body at the current moment by using a pre-established longitudinal acceleration calculation model of the vehicle body corresponding to the current running state of the vehicle and data required by the model; and comparing the expected value of the longitudinal acceleration of the vehicle body at the current moment with the actual value of the longitudinal acceleration of the vehicle body at the current moment, and judging whether the vehicle has a collision accident or not according to the difference between the expected value and the actual value. The invention considers the response delay of the automatic driving control signal, calculates the expected value of the longitudinal acceleration of the vehicle body more accurately, and can realize the effective detection of collision accidents under different running states of the vehicle.
Description
Technical Field
The invention relates to an automatic driving vehicle and a collision accident detection method and system thereof, belonging to the technical field of automatic driving.
Background
With the development of automatic driving technology, the configuration of a safety person on a vehicle is gradually canceled, and finally, L5 level unmanned is realized. Therefore, the vehicle needs to have the capability of autonomously identifying whether a traffic accident occurs, so that the vehicle can stop in time when the traffic accident occurs, loss caused by the accident is avoided by starting again, meanwhile, accident information can be notified to the background, and traffic police can wait for processing, so that accident escape is avoided.
The main form of the traffic accident is a collision accident, and at present, a contact switch or an IMU (Inertial Measurement Unit ) of a vehicle body is generally adopted to judge the collision accident. For example: in the Chinese patent application document with the application number of CN110696766A, a clamping plate mechanism is used for detecting collision, when a collision accident occurs, the double-layer clamping plates are in pressure contact and are conducted, but the method can only detect the condition that the collision point is just provided with the clamping plates, and detection omission easily occurs; in the application document of the Chinese invention with the application number of CN110766982A, a GPS and a triaxial acceleration sensor are used for judging collision accidents, but the method only uses the information obtained by the sensor for collision accident detection, and does not consider road conditions and vehicle running states, and the risk of false detection exists when the vehicle is stopped in a sudden braking manner or the vehicle passes through a special road surface (a deceleration strip, a bumpy road surface, a hollow road surface and the like).
Disclosure of Invention
The invention aims to provide an automatic driving vehicle with high collision accident detection accuracy and a collision accident detection method and system thereof.
In order to achieve the above object, the present invention provides a collision accident detection method of an autonomous vehicle, the method comprising the steps of:
(1) In the running process of an automatic driving vehicle, acquiring the vehicle speed, an automatic driving control signal of the vehicle and the actual value of the longitudinal acceleration of the vehicle body in real time, wherein the automatic driving control signal comprises the opening degree of an accelerator and the braking deceleration;
(2) Determining the running state of the current vehicle according to the vehicle speed, the accelerator opening and the brake deceleration at the current moment, wherein the running state of the vehicle comprises at least one of an acceleration state, a deceleration state and an emergency brake stopping state;
(3) Calculating an expected value of the longitudinal acceleration of the vehicle body at the current moment by using a pre-established longitudinal acceleration calculation model of the vehicle body corresponding to the current running state of the vehicle and data required by the model; the vehicle longitudinal acceleration calculation models corresponding to the acceleration state, the deceleration state and the sudden braking and stopping state are respectively an acceleration model, a deceleration model and a sudden braking and stopping model, data required by the acceleration model comprise the vehicle speed at the current moment and the accelerator opening at the moment t1 before, data required by the deceleration model comprise the braking deceleration at the moment t2 before, and data required by the sudden braking and stopping model comprise the actual value of the vehicle longitudinal acceleration when the acceleration starts to rebound; wherein t1 is acceleration response delay of the vehicle, and t2 is deceleration response delay of the vehicle;
(4) And comparing the expected value of the longitudinal acceleration of the vehicle body at the current moment with the actual value of the longitudinal acceleration of the vehicle body at the current moment, and judging whether the vehicle has a collision accident or not according to the difference between the expected value and the actual value.
The invention also provides a collision accident detection system of an automatic driving vehicle, which comprises:
the whole vehicle controller is used for acquiring the vehicle speed in real time;
The intelligent controller is used for acquiring automatic driving control signals of the vehicle in real time, wherein the automatic driving control signals comprise accelerator opening and braking deceleration;
the acceleration sensor is used for acquiring the actual value of the longitudinal acceleration of the vehicle body in real time;
And the collision accident detection module is used for receiving the data of the whole vehicle controller, the intelligent controller and the acceleration sensor and is used for realizing the collision accident detection method of the automatic driving vehicle.
The invention also provides an autonomous vehicle comprising a vehicle body and a collision accident detection system comprising:
the whole vehicle controller is used for acquiring the vehicle speed in real time;
The intelligent controller is used for acquiring automatic driving control signals of the vehicle in real time, wherein the automatic driving control signals comprise accelerator opening and braking deceleration;
the acceleration sensor is used for acquiring the actual value of the longitudinal acceleration of the vehicle body in real time;
And the collision accident detection module is used for receiving the data of the whole vehicle controller, the intelligent controller and the acceleration sensor and is used for realizing the collision accident detection method of the automatic driving vehicle.
The beneficial effects of the invention are as follows: according to the invention, a corresponding vehicle longitudinal acceleration calculation model is established for each running state of the vehicle, so that the effective detection of collision accidents under different running states of the vehicle can be realized, and false detection is not caused; when the expected value of the longitudinal acceleration of the vehicle body is calculated, the response delay of the automatic driving control signal is fully considered, so that the time alignment of the automatic driving control signal and the longitudinal acceleration signal of the vehicle body is realized, the calculation result of the expected value of the longitudinal acceleration of the vehicle body is more accurate, and the accuracy of collision accident detection is further ensured; in summary, the invention fully considers the running state of the vehicle and the response delay of the automatic driving control signal when detecting the collision accident, and has high collision accident detection accuracy.
Further, in the above-described autonomous vehicle and collision accident detection method and system thereof, the acceleration model is: ay_ expt =k1×a (t-t 1) +k2×v+b1; the deceleration model is: ay_ expt =k3×b (t-t 2) +b2; the emergency brake and parking model is as follows: Where ay_ expt is an expected value of the longitudinal acceleration of the vehicle body, a (t-t 1) is the accelerator opening at time t-t1, V is the vehicle speed at time t, k1, k2, B1 are fitting parameters in an acceleration state, B (t-t 2) is the braking deceleration at time t-t2, k3, B2 are fitting parameters in a deceleration state, a 0 is an actual value of the longitudinal acceleration of the vehicle body when acceleration starts to rebound, a 0 is an acceleration value at the time when acceleration starts to rebound in a typical acceleration rebound model, and a (t) is an acceleration value in a typical acceleration rebound model at time t.
Further, in the above-described autonomous vehicle and collision accident detection method and system thereof, the running state of the vehicle further includes a stationary state, and the current running state of the vehicle is determined by: if the vehicle speed and the braking deceleration at the current moment meet V < k4 x B+b, the current running state of the vehicle is an emergency braking and stopping state; if the current vehicle is not in the sudden braking and stopping state and the vehicle speed is 0, the running state of the current vehicle is a stationary state; if the current vehicle is not in the sudden braking and stopping state and the vehicle speed is not 0, the running state of the current vehicle is in a decelerating state when the braking deceleration is smaller than 0, and the running state of the current vehicle is in an accelerating state when the accelerator opening is larger than 0; in the formula, V is the vehicle speed at the moment t, B is the accelerator opening at the moment t, and k4 and B are fitting parameters.
Further, in the above-mentioned autonomous vehicle and collision accident detection method and system thereof, the acceleration response delay of the vehicle is calculated based on vehicle acceleration data of [0, t ] time period, the vehicle acceleration data being an accelerator opening degree and a longitudinal acceleration of the vehicle body when the accelerator opening degree is greater than 0, the acceleration response delay of the vehicle being such thatΔt at the maximum value of (2); the deceleration response delay of the vehicle is calculated based on vehicle deceleration data in [0, T ] time period, wherein the vehicle deceleration data are braking deceleration and longitudinal acceleration of the vehicle body when the braking deceleration is smaller than 0, and the deceleration response delay of the vehicle is the result/>Δt at the maximum value of (2); where a (t) is the accelerator opening at time t, B (t) is the braking deceleration at time t, and Ay (t+Δt) is the vehicle longitudinal acceleration at time t+Δt.
Further, in the above-mentioned method and system for detecting an automatic driving vehicle and a collision accident thereof, the process of judging whether the vehicle has a collision accident is as follows: calculating a difference value between an expected value and an actual value of the longitudinal acceleration of the vehicle body at the current moment, and judging that the vehicle has a collision accident when the difference value exceeds a collision threshold value; or respectively calculating the difference between the expected value of the longitudinal acceleration of the vehicle body at the current moment and the actual value of the longitudinal acceleration of the vehicle body in a set time period before and after the current moment, taking the difference with the smallest absolute value as the difference between the expected value and the actual value of the longitudinal acceleration of the vehicle body at the current moment, and judging that the vehicle has a collision accident when the difference exceeds a collision threshold value; or respectively solving the difference between the expected value of the longitudinal acceleration of the vehicle body at the current moment and the actual value of the longitudinal acceleration of the vehicle body in a set time period before and after the current moment, respectively comparing each obtained difference with a collision threshold value, and judging that the vehicle has a collision accident when the continuous N differences exceed the collision threshold value, wherein N is more than or equal to 2.
The beneficial effects of doing so are: 3 methods for judging whether collision accidents happen are provided, wherein one of the methods is selected in practical application, and the difference value between the expected value and the actual value of the longitudinal acceleration of the vehicle body at the current moment is directly compared with a collision threshold value, so that the method is the simplest and the most rapid; the method for comparing the difference value with the minimum absolute value with the collision threshold value can solve the detection error caused by the fact that the expected value and the actual value of the longitudinal acceleration of the vehicle body are not completely aligned on a time axis due to small change of response delay of an automatic driving control signal or inaccurate calculation of an acceleration rebound point when the vehicle is stopped, and improves the detection precision of collision accidents; and when the continuous N difference values exceed the collision threshold value, the collision accident of the vehicle is judged, so that the detection error caused by inaccurate actual longitudinal acceleration value of the vehicle body due to signal noise of the acceleration sensor can be solved, and the detection precision of the collision accident is improved.
Further, in the automatic driving vehicle and the collision accident detection method and system thereof, the collision threshold is determined according to the initial threshold and the threshold coefficient corresponding to the current position of the vehicle, the threshold coefficient corresponding to the current position of the vehicle is determined according to the running state of the current vehicle and the road condition of the current position of the vehicle, the road condition comprises a normal road surface and a special road surface, and the special road surface comprises a deceleration strip, a bumpy road surface and a pothole road surface; wherein the threshold coefficient at the normal road surface position is 1, and the threshold coefficient at the special road surface position is greater than 1.
The beneficial effects of doing so are: considering that unexpected changes can occur to the longitudinal acceleration of the vehicle body when the vehicle passes through a deceleration strip, a bumpy road surface, a hollow road surface or other special road surfaces in the normal running (collision-free) process of the vehicle, and unexpected changes can also occur to the longitudinal acceleration of the vehicle body when the vehicle passes through the same special road surface in different running states, the unexpected changes are easily and mistakenly detected as collision accidents, in order to avoid the false detection when the vehicle passes through the special road surface, the collision threshold is related to the running state of the vehicle and the road condition of the current position of the vehicle, the collision threshold at the position of the special road surface is compensated by using the threshold coefficient, the false detection rate when the vehicle passes through the special road surface can be effectively reduced, and the accuracy of the detection of the vehicle collision accidents is ensured.
Further, in the method and the system for detecting the collision accident of the automatic driving vehicle, the threshold coefficient at the position of the special road surface is calibrated through a test, and the calibration process is as follows: the method comprises the steps of respectively measuring actual values of longitudinal acceleration of a vehicle when the vehicle passes through the special road surface position in different running states according to each special road surface position, respectively calculating expected values of longitudinal acceleration of the vehicle when the vehicle passes through the special road surface position in different running states according to real vehicle data obtained when the vehicle passes through the special road surface position in different running states and a vehicle speed longitudinal acceleration calculation model corresponding to the corresponding running states, respectively calculating differences between the expected values and the actual values of the longitudinal acceleration of the vehicle when the vehicle passes through the special road surface position in different running states, and taking the ratio of the differences in each running state to an initial threshold value as a threshold coefficient in the corresponding special road surface position in the corresponding running state.
The beneficial effects of doing so are: considering that the unexpected degree of change of the longitudinal acceleration of the vehicle body is different when the vehicle passes through the same special road surface in different running states, the threshold coefficient of each special road surface position is calibrated for a plurality of times in different running states of the vehicle, so that the threshold coefficient of each special road surface position corresponding to a plurality of different vehicle running states is obtained, the obtained threshold coefficients of the different vehicle running states are utilized to compensate the collision threshold of the special road surface position, the false detection rate of the vehicle passing through the special road surface in different running states can be effectively reduced, and the accuracy of vehicle collision accident detection is ensured.
Further, in the above-mentioned automatic driving vehicle and collision accident detection method and system thereof, the calculation method of the collision threshold corresponding to the current position of the vehicle is as follows: t=kt 0, where T is a collision threshold corresponding to the current position of the vehicle, k is a threshold coefficient corresponding to the current position of the vehicle, k is equal to or greater than 1, and T0 is an initial threshold.
Drawings
FIG. 1 is a schematic view of a collision accident detection system in an embodiment of a vehicle of the present invention;
FIG. 2 is a schematic view of the mounting position and acceleration and angular velocity of a vehicle IMU in a vehicle embodiment of the present invention;
FIG. 3 is a flow chart of a collision accident detection method in an embodiment of the vehicle of the present invention;
FIG. 4 is a schematic illustration of vehicle acceleration response delays and deceleration response delays in an embodiment of a vehicle of the present invention;
FIG. 5 is a schematic diagram of acceleration rebound phenomenon when the vehicle is suddenly braked and stopped in an embodiment of the vehicle according to the invention;
FIG. 6 is a schematic representation of vehicle speed at which acceleration rebound occurs for different braking decelerations in an embodiment of the vehicle of the present invention;
FIG. 7 is a schematic diagram of a typical acceleration rebound model in a vehicle embodiment of the invention;
FIG. 8 is a schematic representation of signal characteristics during normal driving in an embodiment of the vehicle of the present invention;
fig. 9 is a schematic view of signal characteristics in the event of a collision accident in an embodiment of the vehicle of the present invention.
Detailed Description
The present invention will be described in further detail with reference to the drawings and examples, in order to make the objects, technical solutions and advantages of the present invention more apparent.
Vehicle embodiment:
The autonomous vehicle of the present embodiment includes a vehicle body and a collision accident detection system, as shown in fig. 1, the collision accident detection system includes: the system comprises a whole vehicle controller, an intelligent controller, an acceleration sensor (such as a triaxial acceleration sensor or a biaxial acceleration sensor), a high-precision map module and a collision accident detection module, wherein the collision accident detection module comprises a vehicle body longitudinal acceleration calculation module, a collision accident logic judgment module and a collision accident judgment result output module.
The vehicle control system comprises a vehicle controller, an acceleration sensor, a high-precision map module and a high-precision map module, wherein the vehicle controller is used for acquiring the vehicle speed in real time, the intelligent controller is used for acquiring an automatic driving control signal (comprising the opening degree of an accelerator and the braking deceleration) of the vehicle in real time, the acceleration sensor is used for acquiring the actual value of the longitudinal acceleration of the vehicle body in real time, the high-precision map module stores a high-precision map acquired in advance, special roads (such as a deceleration strip, a bumpy road surface, a pothole road surface and the like) are marked in the high-precision map, the threshold value coefficient corresponding to the corresponding special road surface is marked at each special road surface position in the high-precision map, and the high-precision map module is used for providing the threshold value coefficient corresponding to the current position of the vehicle; the collision accident detection module is used for receiving the vehicle speed, the automatic driving control signal of the vehicle, the actual value of the longitudinal acceleration of the vehicle body and the threshold coefficient corresponding to the current position of the vehicle, and is used for realizing the collision accident detection method shown in fig. 3. Specifically, the vehicle longitudinal acceleration calculation module is used for determining the running state of the current vehicle according to the vehicle speed, the accelerator opening and the brake deceleration at the current moment, and calculating to obtain the expected value of the vehicle longitudinal acceleration at the current moment by utilizing a pre-established vehicle longitudinal acceleration calculation model corresponding to the running state of the current vehicle and data required by the model; the collision accident logic judging module is used for determining a collision threshold value corresponding to the current position of the vehicle according to the initial threshold value and a threshold value coefficient corresponding to the current position of the vehicle, comparing an expected value of the longitudinal acceleration of the vehicle body at the current moment with an actual value of the longitudinal acceleration of the vehicle body at the current moment according to the collision threshold value corresponding to the current position of the vehicle, and judging whether the vehicle has a collision accident or not according to the difference between the expected value and the actual value of the longitudinal acceleration of the vehicle body at the current moment; the collision accident judgment result output module is used for outputting the detection result of the vehicle collision accident.
In this embodiment, a triaxial acceleration sensor IMU is selected as an acceleration sensor, as shown in fig. 2, the IMU is mounted on the top of the vehicle to provide acceleration and angular velocity in three directions of the vehicle, where Ay is longitudinal acceleration of the vehicle body, Ω y is longitudinal angular velocity of the vehicle body, ax is transverse acceleration of the vehicle body, Ω x is transverse angular velocity of the vehicle body, az is vertical acceleration of the vehicle body, and Ω z is vertical angular velocity of the vehicle body.
The collision accident detection method of the embodiment is a collision accident detection method based on multi-source information fusion of vehicle acceleration sensor information, vehicle running state, automatic driving control signals of a vehicle and high-precision map information, and the specific implementation process of the method is shown in fig. 3:
Step 1, acquiring the actual values of the vehicle speed, an automatic driving control signal (comprising accelerator opening and brake deceleration) of the vehicle and the longitudinal acceleration of the vehicle body in real time in the running process of the automatic driving vehicle;
Step 2, determining the running state of the current vehicle according to the vehicle speed, the accelerator opening and the brake deceleration at the current moment, wherein the running state of the vehicle comprises an acceleration state, a deceleration state, an emergency brake stopping state and a static state; the current vehicle operating state is determined by the following method: if the vehicle speed and the braking deceleration at the current moment meet the acceleration rebound condition, the running state of the current vehicle is an emergency braking and stopping state, wherein the acceleration rebound condition is met when a formula V < k4 x B+b is met, V is the vehicle speed at the moment t, B is the accelerator opening at the moment t, and k4 and B are fitting parameters; if the current vehicle is not in the sudden braking and stopping state and the vehicle speed is 0, the running state of the current vehicle is a stationary state; if the current vehicle is not in the sudden braking and stopping state and the vehicle speed is not 0, the running state of the current vehicle is in a decelerating state when the braking deceleration is smaller than 0, and the running state of the current vehicle is in an accelerating state when the accelerator opening is larger than 0.
Step 3, calculating to obtain an expected value Ay_Expt of the longitudinal acceleration of the vehicle body at the current moment by using a pre-established longitudinal acceleration calculation model of the vehicle body corresponding to the current vehicle running state and data required by the model;
The vehicle longitudinal acceleration calculation models corresponding to each vehicle running state are different, and the vehicle longitudinal acceleration calculation models corresponding to the acceleration state, the deceleration state, the sudden braking and stopping state and the static state are respectively an acceleration model, a deceleration model, a sudden braking and stopping model and a static model. Each model is described in detail below:
The acceleration model is: ay_ expt =k1.A (t-t 1) +k2.V+b1, wherein A (t-t 1) is the accelerator opening at the time of t-t1, t1 is the acceleration response delay of the vehicle, V is the vehicle speed at the time of t, and k1, k2 and b1 are fitting parameters in an acceleration state; the data required by the acceleration model can be seen to be the vehicle speed at the current moment and the accelerator opening at the previous moment t 1;
The deceleration model is: ay_ expt =k3+B2, wherein B (t-t 2) is the braking deceleration at the moment of t-t2, t2 is the deceleration response delay of the vehicle, and k3 and B2 are fitting parameters in a deceleration state; the data required by the deceleration model can be seen to be the braking deceleration at the previous time t 2;
The emergency brake and parking model is as follows: Wherein A 0 is the actual value of the longitudinal acceleration of the vehicle body when the acceleration starts to rebound, a 0 is the acceleration value when the acceleration starts to rebound in a typical acceleration rebound model, and a (t) is the acceleration value in the typical acceleration rebound model at the time t; the data required by the sudden braking and stopping model can be seen to be the actual value of the longitudinal acceleration of the vehicle body when the acceleration starts to rebound, the acceleration value when the acceleration starts to rebound in the typical acceleration rebound model and the acceleration value in the typical acceleration rebound model at the current moment;
the stationary model is: ay_ expt =0.
The establishment process of the vehicle longitudinal acceleration calculation model corresponding to each vehicle running state is as follows:
1) Calculating response delays (including acceleration response delays and deceleration response delays) of the vehicle;
Since the vehicle responds to the acceleration and deceleration actions from the vehicle autopilot control signals (i.e., the accelerator opening control signal and the brake deceleration control signal) and then the acceleration sensor detects the state change of the vehicle, there is a certain time delay (this time is called a response delay) in the period, so in order to establish a mathematical relationship model (i.e., a vehicle longitudinal acceleration calculation model) between the vehicle acceleration and the vehicle state and the vehicle autopilot control signal, the response delay of the vehicle needs to be calculated first.
As shown in fig. 4, the horizontal axis in the figure represents time (unit: seconds), the vertical axis represents vehicle longitudinal acceleration Ay (unit: m/s 2), accelerator opening a (unit: degree, which is scaled down 40 times for convenience of display), brake deceleration B (unit: m/s 2), vehicle speed V (unit: km/h, which is scaled down 10 times for convenience of display), t1 is a time interval for transmitting an accelerator opening control signal to acceleration of a vehicle response, and t2 is a time interval for transmitting a brake deceleration control signal to deceleration of a vehicle response, and it can be seen that response delay at acceleration and response delay at deceleration are different.
Since the response delay at acceleration and the response delay at deceleration are different, in order to obtain the acceleration response delay and the deceleration response delay, it is necessary to separate data at acceleration and data at deceleration and then calculate the data respectively, and the calculation methods are as follows:
And taking vehicle acceleration data in the [0, T ] time period, wherein the vehicle acceleration data is the accelerator opening degree and the longitudinal acceleration of the vehicle body when the accelerator opening degree is more than 0, the vehicle acceleration data is calculated according to the formula (1), and the acceleration response delay is a delta t value when the value of the formula (1) is maximized.
And taking vehicle deceleration data in the [0, T ] time period, wherein the vehicle deceleration data is the braking deceleration and the longitudinal acceleration of the vehicle body when the braking deceleration is smaller than 0, and calculating according to the formula (2), and the deceleration response delay is the delta t value when the value of the formula (2) is maximum.
Where a (t) is the accelerator opening at time t, B (t) is the braking deceleration at time t, and Ay (t+Δt) is the vehicle longitudinal acceleration at time t+Δt.
2) Fitting to obtain a vehicle body longitudinal acceleration calculation model of each running state of the vehicle based on response delay of the vehicle and a large amount of collected real vehicle data when the vehicle runs normally; wherein, through data analysis, divide the vehicle running state into: the vehicle longitudinal acceleration calculation model is different when the vehicle is in different running states.
The vehicle longitudinal acceleration calculation model of each vehicle running state is as follows:
(1) Acceleration model
When the vehicle is accelerated normally, the acceleration is a positive value, the magnitude of the acceleration increases along with the increase of the accelerator opening, but the acceleration value tends to be gentle along with the increase of the vehicle speed. Therefore, when the vehicle accelerates, the magnitude of the acceleration is related to the accelerator opening and the vehicle speed, and the acceleration calculation model when the vehicle is in an accelerating state is as follows:
Ay_expt＝k1*A(t-t1)+k2*V+b1 (3)
wherein A is the accelerator opening degree, t1 is the acceleration response delay, and k1, k2 and b1 are fitting parameters during acceleration.
In order to acquire fitting parameters during acceleration, automatic driving data for a period of time need to be acquired in advance, data only comprising an acceleration process is extracted, a relation between vehicle acceleration and vehicle speed and accelerator opening is established by using a formula (3), and fitting parameters are obtained by using a least square fitting method.
(2) Deceleration model
When the vehicle is decelerating normally, the acceleration is negative, the control signal directly issues a braking deceleration value to the brake, but the vehicle body response and control issues are not always exactly identical. Through data analysis, the actual acceleration of the vehicle body and the braking deceleration issued by the control signal are positively correlated, and the acceleration calculation model when the vehicle is in a deceleration state is as follows:
Ay_expt＝k3*B(t-t2)+b2 (4)
Wherein B is braking deceleration, t2 is deceleration response delay, k3 and B2 are fitting parameters during deceleration, and the calculation method is consistent with the calculation method of the fitting parameters during acceleration.
(3) Emergency braking and stopping model
When the vehicle is suddenly braked and parked, the vehicle body swings back and forth, and the change rule of the acceleration value and the acceleration value in a normal braking state is large in difference, so that calculation cannot be performed according to the formula (4). The vehicle body and the chassis are connected through the suspension, which is equivalent to a spring structure, when the vehicle is suddenly braked, the kinetic energy of the vehicle body is converted into elastic potential energy, and when the vehicle is stopped or approaches to a stop, the elastic potential energy is released and converted into the kinetic energy of the vehicle body, so that the vehicle body swings back and forth. The change law of the acceleration of the vehicle body at this time can be observed by the data change measured by the acceleration sensor, as shown in fig. 5.
As can be seen in FIG. 5, when the vehicle is just started to brake (3 s-4 s), the vehicle acceleration and the brake deceleration are positively correlated and gradually reach about-3 m/s2, and the vehicle speed gradually decreases at this time; the acceleration of the vehicle is kept unchanged at about-3 m/s2, and the vehicle is still in a decelerating state (4 s-6 s); when the vehicle speed approaches 0, the acceleration curve starts to rise, and then periodic oscillation occurs, the amplitude gradually decreases, and finally approaches 0 (6 s-8 s); the position at which acceleration starts to rise is called the acceleration rebound point.
Through data analysis, a certain relation exists between the speed of the vehicle when the acceleration rebounds and the magnitude of the braking deceleration, the speed of the vehicle when the acceleration rebounds under different braking decelerations is counted, the result is shown as blue dispersion points in fig. 6, the dispersion points are fitted, and the relation between the speed of the vehicle and the braking deceleration when the acceleration rebounds can be obtained, as shown as red straight lines in the figure. Therefore, when the vehicle speed satisfies the formula (5), the acceleration starts to rebound:
V<k4*B+b (5)
Meanwhile, the frequency of the acceleration vibration is fixed (related to the natural frequency of the vehicle suspension) through data analysis, and the amplitude is only related to the acceleration when the acceleration starts to rebound. Therefore, a typical acceleration rebound model (as shown in fig. 7) is formed by collecting a section of acceleration oscillation signal in advance, and in the actual sudden braking process, the acceleration value ay_ expt expected at the corresponding moment can be obtained only by scaling the typical acceleration rebound model.
If the acceleration value at which rebound actually starts is a 0, the acceleration ay_ expt during rebound can be expressed as:
Where a 0 denotes the acceleration at which rebound starts in a typical acceleration rebound model, For scaling, a (t) is the acceleration at time t in a typical acceleration rebound model.
(4) Stationary model
When the vehicle is stationary, the vehicle acceleration value is 0, namely:
Ay_expt＝0 (7)
Because the performance of each automatic driving vehicle is not completely consistent, the response delay time of each vehicle and the parameters in the vehicle body acceleration calculation model corresponding to each running state are required to be calibrated, the calculation accuracy of the expected result is ensured, and the risk of false detection is reduced.
And 4, comparing the expected value of the longitudinal acceleration of the vehicle body at the current moment with the actual value of the longitudinal acceleration of the vehicle body at the current moment, and judging whether the vehicle has a collision accident or not according to the difference between the expected value and the actual value.
The following provides 3 methods for judging whether the vehicle has a collision accident, and one of the methods can be selected in practical application:
method (1): calculating a difference value between an expected value and an actual value of the longitudinal acceleration of the vehicle body at the current moment, and judging that the vehicle has a collision accident when the difference value exceeds a collision threshold value corresponding to the current position of the vehicle; the method is the simplest and most rapid.
Method (2): respectively calculating difference values of the expected value of the longitudinal acceleration of the vehicle body at the current moment and the actual value of the longitudinal acceleration of the vehicle body in a set time period before and after the current moment, taking the difference value with the smallest absolute value as the difference value between the expected value and the actual value of the longitudinal acceleration of the vehicle body at the current moment, and judging that the vehicle has a collision accident when the difference value exceeds a collision threshold value corresponding to the current position of the vehicle; the length of the time period set before and after the current time can be determined according to actual needs, for example, taking the actual value of the longitudinal acceleration of the vehicle body between 20ms before the current time and 20ms after the current time (if the current time is the time t, namely taking the actual value of the longitudinal acceleration of the vehicle body between t-20ms and t+20ms), or taking the actual value of the longitudinal acceleration of the vehicle body between the previous frame and the subsequent frame of the current time (namely taking the actual value of the longitudinal acceleration of the vehicle body between the previous 1 frame, the current frame and the subsequent 1 frame of the current time); the method can solve the detection error caused by the fact that the expected value and the actual value of the longitudinal acceleration of the vehicle body are not completely aligned on a time axis due to the small change of response delay of an automatic driving control signal or inaccurate calculation of an acceleration rebound point during sudden braking and stopping, and improves the detection precision of collision accidents;
method (3): calculating difference values of the expected value of the longitudinal acceleration of the vehicle body at the current moment and the actual value of the longitudinal acceleration of the vehicle body in a set time period (same as that in the method (2)) before and after the current moment, comparing each calculated difference value with a collision threshold value respectively, and judging that the vehicle has a collision accident when the continuous N difference values exceed the collision threshold value corresponding to the current position of the vehicle, wherein N is more than or equal to 2; the method can solve the detection error caused by inaccurate actual longitudinal acceleration value of the vehicle body due to signal noise of the acceleration sensor, and improves the detection precision of collision accidents.
As shown in fig. 3, the method (3) is selected to detect a collision accident in the present embodiment, and when the continuous 2 differences all exceed the collision threshold corresponding to the current position of the vehicle, the collision accident of the vehicle is determined.
The method for calculating the collision threshold T corresponding to the current position of the vehicle comprises the following steps: t=k×t0, where k is a threshold coefficient corresponding to the current position of the vehicle, and T0 is an initial threshold; the threshold coefficient k and the initial threshold T0 corresponding to the current position of the vehicle are obtained through test calibration.
The threshold coefficient k corresponding to the current position of the vehicle is determined according to the running state of the current vehicle and the road condition of the current position of the vehicle, the road condition comprises a normal road surface and a special road surface (comprising a deceleration strip, a bumpy road surface and a hollow road surface), wherein the threshold coefficient at the position of the normal road surface is 1, the threshold coefficient at the position of the special road surface is greater than 1, namely when the current position of the vehicle is the normal road surface, k=1, and when the current position of the vehicle is the special road surface, k is more than 1.
The advantage of this is: considering that unexpected changes can occur to the longitudinal acceleration of the vehicle body when the vehicle passes through a deceleration strip, a bumpy road surface, a hollow road surface or other special road surfaces in the normal running (collision-free) process of the vehicle, and unexpected changes can also occur to the longitudinal acceleration of the vehicle body when the vehicle passes through the same special road surface in different running states, the unexpected changes are easily and mistakenly detected as collision accidents, in order to avoid the false detection when the vehicle passes through the special road surface, the collision threshold is related to the running state of the vehicle and the road condition of the current position of the vehicle, the collision threshold at the position of the special road surface is compensated by using the threshold coefficient, the false detection rate when the vehicle passes through the special road surface can be effectively reduced, and the accuracy of the detection of the vehicle collision accidents is ensured.
The threshold coefficient at the position of the special road surface is calibrated through a test, and the calibration process is as follows: for each special road surface position, respectively measuring actual values of longitudinal acceleration of the vehicle when the vehicle passes through the special road surface position in different running states, respectively calculating expected values of longitudinal acceleration of the vehicle when the vehicle passes through the special road surface position in different running states based on acquired real vehicle data of the vehicle passing through the special road surface position in different running states and a vehicle speed longitudinal acceleration calculation model corresponding to the corresponding running states, respectively calculating differences between the expected values and the actual values of the longitudinal acceleration of the vehicle corresponding to the special road surface position in each running state, and taking the ratio of the differences to an initial threshold value in each running state as a threshold value coefficient in the special road surface position in the corresponding running state; that is, for the same special road surface position, it is necessary to calibrate the threshold value coefficient at the special road surface position a plurality of times under different running states of the vehicle, so that the same special road surface position corresponds to a plurality of threshold value coefficients, and each threshold value coefficient corresponds to one running state of the vehicle. For example: the threshold value coefficient at a specific road surface position is calibrated when the vehicle is in an accelerating state, and the threshold value coefficient at the specific road surface position when the vehicle is in the accelerating state is obtained, and in the embodiment, 4 vehicle running states exist, and the threshold value coefficient at the specific road surface position corresponds to the vehicle running states.
When the collision accident is detected, when the vehicle passes through a normal road surface, the collision threshold corresponding to the current position of the vehicle is an initial threshold; when the vehicle passes through a certain special road surface in an acceleration state, a collision threshold value corresponding to the current position of the vehicle=a threshold value coefficient x an initial threshold value at the position of the special road surface when the vehicle is in the acceleration state, and similarly, when the vehicle passes through a certain special road surface in other running states, a collision threshold value corresponding to the current position of the vehicle=a threshold value coefficient x an initial threshold value at the position of the special road surface when the vehicle is in the corresponding state.
The advantage of this is: considering that the unexpected degree of change of the longitudinal acceleration of the vehicle body is different when the vehicle passes through the same special road surface in different running states, the threshold coefficient of each special road surface position is calibrated for a plurality of times in different running states of the vehicle, so that the threshold coefficient of each special road surface position corresponding to a plurality of different vehicle running states is obtained, the obtained threshold coefficients of the different vehicle running states are utilized to compensate the collision threshold of the special road surface position, the false detection rate of the vehicle passing through the special road surface in different running states can be effectively reduced, and the accuracy of vehicle collision accident detection is ensured.
In the present embodiment, a specific manner of determining the threshold value coefficient in the form of the ratio of the difference value between the expected value and the actual value of the longitudinal acceleration of the vehicle body to the initial threshold value is given, and as other embodiments, the threshold value coefficient may be determined in other manners than the ratio; in this embodiment, considering that there is a special road surface in the operation scene of the automatic driving vehicle, the threshold coefficient is set to compensate the influence of the special road surface on the collision accident detection result, and as other embodiments, when there is only a normal road surface in the operation scene of the automatic driving vehicle and no special road surface exists, the threshold coefficient may not be set, and at this time, the collision threshold value at each position of the vehicle is the initial threshold value, and the collision accident detection is performed by using the initial threshold value.
The running state of the vehicle in this embodiment includes an acceleration state, a deceleration state, an abrupt brake stop state, and a stationary state, and as other embodiments, the running state of the vehicle may include at least one of an acceleration state, a deceleration state, and an abrupt brake stop state.
In the present embodiment, the vehicle longitudinal acceleration calculation model takes as input an automatic driving control signal (accelerator opening and brake deceleration), that is, takes as normal change a vehicle acceleration change caused by the automatic driving control signal, and takes as a collision accident a vehicle acceleration change caused by a non-automatic driving control signal. In order to check the effectiveness of the collision accident detection method of the present embodiment, a debugging manual pressing of the brake pedal may be used to perform manual intervention, and unexpected deceleration of the vehicle due to a collision may be simulated. The method has the advantages that signals simulating the collision can be acquired without performing a collision experiment, and meanwhile, the verification of the collision accident detection method can be performed in a collision simulation mode.
As shown in fig. 8, fig. 8 shows the signal variation characteristics of the vehicle during normal running, and includes an actual acceleration curve, an expected acceleration curve, and a calculated difference S between the expected value and the actual value. As can be seen, the difference S in the whole process is within 1m/S2, so that no collision accident can be judged; as shown in fig. 9, fig. 9 is a signal change characteristic at the time of a collision accident, in 3S-4S, the expected acceleration of the vehicle becomes 0, and the actual acceleration suddenly decreases, so that it is considered that the vehicle is decelerated due to the collision, the difference S between the expected value and the actual value of the acceleration suddenly increases, and the collision accident is judged to occur beyond the initial threshold t0=1 m/S2. Here, fig. 8 and 9 are collision accident detection under a non-special road surface, and thus the collision threshold initial threshold t0=1 m/s2.
The initial threshold t0=1m/s 2 is the result of a number of data tests, in which the measurement error of the acceleration sensor and the fitting calculation error are taken into account. The meaning is as follows: based on the accuracy and algorithm of the current acceleration sensor, a collision accident that causes the acceleration of the own vehicle to vary by more than 1m/s2 can be detected. As other embodiments, the high-precision acceleration sensor can be used, the measurement error of acceleration is reduced, the acceleration signal fitting and processing algorithm is optimized, and the interference caused by signal noise is reduced, so that the accuracy of collision accident detection is improved, and a more slight collision accident can be detected.
The collision accident detection method of the embodiment has the beneficial effects that:
(1) Calculating response delay of the automatic driving control signal, realizing time alignment of the automatic driving control signal and the vehicle body longitudinal acceleration signal, and calculating a vehicle body longitudinal acceleration expected value more accurately;
(2) Corresponding vehicle longitudinal acceleration calculation models are established for each running state of the vehicle, so that the effective detection of collision accidents under different running states of the vehicle can be realized, and false detection is avoided;
(3) The threshold value coefficient at the position of the special road surface is marked, and the collision threshold value at the position of the special road surface is compensated by using the threshold value coefficient, so that no false detection and no missed detection of collision accidents can be realized on the special road surface;
(4) The signal acquisition of the collision accident and the test verification of the collision accident detection method can be completed under the condition that no real collision occurs.
The collision accident detection method of the embodiment can detect the collision of the self-vehicle or the rear-end collision of other vehicles and cause the collision accident that the acceleration of the self-vehicle changes to a certain extent. The method is not limited to a single vehicle type, and can be applied to all vehicles capable of acquiring vehicle control signals, vehicle body state signals and acceleration signals.
System embodiment:
the collision accident detection system of the present embodiment is the same as that of the vehicle embodiment, and will not be described here again.
Method embodiment:
The collision accident detection method of the present embodiment is the same as that of the vehicle embodiment, and will not be described here again.
Claims (10)
1. A collision accident detection method for an autonomous vehicle, the method comprising the steps of:
(1) In the running process of an automatic driving vehicle, acquiring the vehicle speed, an automatic driving control signal of the vehicle and the actual value of the longitudinal acceleration of the vehicle body in real time, wherein the automatic driving control signal comprises the opening degree of an accelerator and the braking deceleration;
(2) Determining the running state of the current vehicle according to the vehicle speed, the accelerator opening and the brake deceleration at the current moment, wherein the running state of the vehicle comprises at least one of an acceleration state, a deceleration state and an emergency brake stopping state;
(3) Calculating an expected value of the longitudinal acceleration of the vehicle body at the current moment by using a pre-established longitudinal acceleration calculation model of the vehicle body corresponding to the current running state of the vehicle and data required by the model; the vehicle longitudinal acceleration calculation models corresponding to the acceleration state, the deceleration state and the sudden braking and stopping state are respectively an acceleration model, a deceleration model and a sudden braking and stopping model, data required by the acceleration model comprise the vehicle speed at the current moment and the accelerator opening at the moment t1 before, data required by the deceleration model comprise the braking deceleration at the moment t2 before, and data required by the sudden braking and stopping model comprise the actual value of the vehicle longitudinal acceleration when the acceleration starts to rebound; wherein t1 is acceleration response delay of the vehicle, and t2 is deceleration response delay of the vehicle;
(4) And comparing the expected value of the longitudinal acceleration of the vehicle body at the current moment with the actual value of the longitudinal acceleration of the vehicle body at the current moment, and judging whether the vehicle has a collision accident or not according to the difference between the expected value and the actual value.
2. The collision accident detection method of an automatically driven vehicle according to claim 1, wherein the acceleration model is: ay_ expt =k1×a (t-t 1) +k2×v+b1; the deceleration model is: ay_ expt =k3×b (t-t 2) +b2; the emergency brake and parking model is as follows: Where ay_ expt is an expected value of the longitudinal acceleration of the vehicle body, a (t-t 1) is the accelerator opening at time t-t1, V is the vehicle speed at time t, k1, k2, B1 are fitting parameters in an acceleration state, B (t-t 2) is the braking deceleration at time t-t2, k3, B2 are fitting parameters in a deceleration state, a 0 is an actual value of the longitudinal acceleration of the vehicle body when acceleration starts to rebound, a 0 is an acceleration value at the time when acceleration starts to rebound in a typical acceleration rebound model, and a (t) is an acceleration value in a typical acceleration rebound model at time t.
3. The collision accident detection method of an autonomous vehicle according to claim 2, wherein the running state of the vehicle further includes a stationary state, and the current running state of the vehicle is determined by: if the vehicle speed and the braking deceleration at the current moment meet V < k4 x B+b, the current running state of the vehicle is an emergency braking and stopping state; if the current vehicle is not in the sudden braking and stopping state and the vehicle speed is 0, the running state of the current vehicle is a stationary state; if the current vehicle is not in the sudden braking and stopping state and the vehicle speed is not 0, the running state of the current vehicle is in a decelerating state when the braking deceleration is smaller than 0, and the running state of the current vehicle is in an accelerating state when the accelerator opening is larger than 0; in the formula, V is the vehicle speed at the moment t, B is the accelerator opening at the moment t, and k4 and B are fitting parameters.
4. The method for detecting a collision accident of an automatically driven vehicle according to claim 3, wherein the acceleration response delay of the vehicle is calculated based on vehicle acceleration data of [0, t ] time period, the vehicle acceleration data being an accelerator opening degree and a longitudinal acceleration of the vehicle when the accelerator opening degree is greater than 0, the acceleration response delay of the vehicle being such that Δt at the maximum value of (2); the deceleration response delay of the vehicle is calculated based on vehicle deceleration data in [0, T ] time period, wherein the vehicle deceleration data are braking deceleration and longitudinal acceleration of the vehicle body when the braking deceleration is smaller than 0, and the deceleration response delay of the vehicle is the result/>Δt at the maximum value of (2); where a (t) is the accelerator opening at time t, B (t) is the braking deceleration at time t, and Ay (t+Δt) is the vehicle longitudinal acceleration at time t+Δt.
5. The collision accident detection method for an automatically driven vehicle according to any one of claims 1 to 4, wherein the process of judging whether the collision accident occurs in the vehicle is: calculating a difference value between an expected value and an actual value of the longitudinal acceleration of the vehicle body at the current moment, and judging that the vehicle has a collision accident when the difference value exceeds a collision threshold value; or respectively calculating the difference between the expected value of the longitudinal acceleration of the vehicle body at the current moment and the actual value of the longitudinal acceleration of the vehicle body in a set time period before and after the current moment, taking the difference with the smallest absolute value as the difference between the expected value and the actual value of the longitudinal acceleration of the vehicle body at the current moment, and judging that the vehicle has a collision accident when the difference exceeds a collision threshold value; or respectively solving the difference between the expected value of the longitudinal acceleration of the vehicle body at the current moment and the actual value of the longitudinal acceleration of the vehicle body in a set time period before and after the current moment, respectively comparing each obtained difference with a collision threshold value, and judging that the vehicle has a collision accident when the continuous N differences exceed the collision threshold value, wherein N is more than or equal to 2.
6. The method for detecting a collision accident of an automatically driven vehicle according to claim 5, wherein the collision threshold is determined according to an initial threshold and a threshold coefficient corresponding to a current position of the vehicle, the threshold coefficient corresponding to the current position of the vehicle is determined according to a running state of the current vehicle and a road condition of the current position of the vehicle, the road condition comprises a normal road surface and a special road surface, and the special road surface comprises a deceleration strip, a bumpy road surface and a pothole road surface; wherein the threshold coefficient at the normal road surface position is 1, and the threshold coefficient at the special road surface position is greater than 1.
7. The method for detecting the collision accident of the automatically driven vehicle according to claim 6, wherein the threshold coefficient at the special road surface position is calibrated through a test, and the calibration process is as follows: the method comprises the steps of respectively measuring actual values of longitudinal acceleration of a vehicle when the vehicle passes through the special road surface position in different running states according to each special road surface position, respectively calculating expected values of longitudinal acceleration of the vehicle when the vehicle passes through the special road surface position in different running states according to real vehicle data obtained when the vehicle passes through the special road surface position in different running states and a vehicle speed longitudinal acceleration calculation model corresponding to the corresponding running states, respectively calculating differences between the expected values and the actual values of the longitudinal acceleration of the vehicle when the vehicle passes through the special road surface position in different running states, and taking the ratio of the differences in each running state to an initial threshold value as a threshold coefficient in the corresponding special road surface position in the corresponding running state.
8. The method for detecting a collision accident of an automatically driven vehicle according to claim 7, wherein the calculation method of the collision threshold corresponding to the current position of the vehicle is: t=kt 0, where T is a collision threshold corresponding to the current position of the vehicle, k is a threshold coefficient corresponding to the current position of the vehicle, k is equal to or greater than 1, and T0 is an initial threshold.
9. A collision accident detection system for an autonomous vehicle, the system comprising:
the whole vehicle controller is used for acquiring the vehicle speed in real time;
The intelligent controller is used for acquiring automatic driving control signals of the vehicle in real time, wherein the automatic driving control signals comprise accelerator opening and braking deceleration;
the acceleration sensor is used for acquiring the actual value of the longitudinal acceleration of the vehicle body in real time;
And a collision accident detection module for receiving data of the whole vehicle controller, the intelligent controller and the acceleration sensor, and for implementing the collision accident detection method of the autonomous vehicle according to any one of claims 1 to 8.
10. An autonomous vehicle comprising a vehicle body and a collision accident detection system, the collision accident detection system comprising:
the whole vehicle controller is used for acquiring the vehicle speed in real time;
The intelligent controller is used for acquiring automatic driving control signals of the vehicle in real time, wherein the automatic driving control signals comprise accelerator opening and braking deceleration;
the acceleration sensor is used for acquiring the actual value of the longitudinal acceleration of the vehicle body in real time;
And a collision accident detection module for receiving data of the whole vehicle controller, the intelligent controller and the acceleration sensor, and for implementing the collision accident detection method of the autonomous vehicle according to any one of claims 1 to 8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011487872.7A CN114633743B (en) | 2020-12-16 | 2020-12-16 | Automatic driving vehicle and collision accident detection method and system thereof |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011487872.7A CN114633743B (en) | 2020-12-16 | 2020-12-16 | Automatic driving vehicle and collision accident detection method and system thereof |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114633743A CN114633743A (en) | 2022-06-17 |
| CN114633743B true CN114633743B (en) | 2024-05-28 |
Family
ID=81945450
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011487872.7A Active CN114633743B (en) | 2020-12-16 | 2020-12-16 | Automatic driving vehicle and collision accident detection method and system thereof |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114633743B (en) |
Citations (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6233515B1 (en) * | 1998-12-07 | 2001-05-15 | Jaguar Car, Limited | Adaptive vehicle cruise control system and methodology |
| CN102616235A (en) * | 2012-04-09 | 2012-08-01 | 北京航空航天大学 | Cooperative anti-collision device based on vehicle-vehicle communication and anti-collision method |
| KR20140140845A (en) * | 2013-05-30 | 2014-12-10 | 서강대학교산학협력단 | Method of characterizing driver propensity for different forward collision warning times |
| CN105493165A (en) * | 2013-09-02 | 2016-04-13 | 丰田自动车株式会社 | Vehicle driving situation determination apparatus and vehicle driving situation determination method |
| CN106891846A (en) * | 2016-12-30 | 2017-06-27 | 上海蔚来汽车有限公司 | Safety warning system and method based on Doppler effect |
| CN106934876A (en) * | 2017-03-16 | 2017-07-07 | 广东翼卡车联网服务有限公司 | A kind of recognition methods of vehicle abnormality driving event and system |
| CN107264527A (en) * | 2017-06-08 | 2017-10-20 | 广州汽车集团股份有限公司 | Intelligent vehicle prevents the control method and device of other car |
| CN109117709A (en) * | 2017-06-23 | 2019-01-01 | 优步技术公司 | Collision avoidance system for automatic driving vehicle |
| JP2019139331A (en) * | 2018-02-06 | 2019-08-22 | パナソニックＩｐマネジメント株式会社 | Vehicle management system and control method of vehicle management system |
| CN110435647A (en) * | 2019-07-26 | 2019-11-12 | 大连理工大学 | A kind of vehicle safety anticollision control method of the TTC based on rolling optimization parameter |
| WO2020082778A1 (en) * | 2018-10-25 | 2020-04-30 | 广州小鹏汽车科技有限公司 | Vehicle collision detection method and vehicle control system |
| WO2020187259A1 (en) * | 2019-03-18 | 2020-09-24 | 长城汽车股份有限公司 | Safety monitoring method and system for autonomous vehicle, and motion control system |
| WO2020187254A1 (en) * | 2019-03-18 | 2020-09-24 | 长城汽车股份有限公司 | Longitudinal control method and system for automatic driving vehicle |
| CN111746555A (en) * | 2019-03-26 | 2020-10-09 | 通用汽车环球科技运作有限责任公司 | Identification and avoidance of collision behavior |
| CN111976638A (en) * | 2020-08-28 | 2020-11-24 | 广州市网优优信息技术开发有限公司 | Vehicle collision detection method and system based on vehicle-mounted intelligent terminal |
-
2020
- 2020-12-16 CN CN202011487872.7A patent/CN114633743B/en active Active
Patent Citations (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6233515B1 (en) * | 1998-12-07 | 2001-05-15 | Jaguar Car, Limited | Adaptive vehicle cruise control system and methodology |
| CN102616235A (en) * | 2012-04-09 | 2012-08-01 | 北京航空航天大学 | Cooperative anti-collision device based on vehicle-vehicle communication and anti-collision method |
| KR20140140845A (en) * | 2013-05-30 | 2014-12-10 | 서강대학교산학협력단 | Method of characterizing driver propensity for different forward collision warning times |
| CN105493165A (en) * | 2013-09-02 | 2016-04-13 | 丰田自动车株式会社 | Vehicle driving situation determination apparatus and vehicle driving situation determination method |
| CN106891846A (en) * | 2016-12-30 | 2017-06-27 | 上海蔚来汽车有限公司 | Safety warning system and method based on Doppler effect |
| WO2018120831A1 (en) * | 2016-12-30 | 2018-07-05 | 上海蔚来汽车有限公司 | Doppler effect-based vehicle safety forewarning system and method |
| CN106934876A (en) * | 2017-03-16 | 2017-07-07 | 广东翼卡车联网服务有限公司 | A kind of recognition methods of vehicle abnormality driving event and system |
| CN107264527A (en) * | 2017-06-08 | 2017-10-20 | 广州汽车集团股份有限公司 | Intelligent vehicle prevents the control method and device of other car |
| CN109117709A (en) * | 2017-06-23 | 2019-01-01 | 优步技术公司 | Collision avoidance system for automatic driving vehicle |
| JP2019139331A (en) * | 2018-02-06 | 2019-08-22 | パナソニックＩｐマネジメント株式会社 | Vehicle management system and control method of vehicle management system |
| WO2020082778A1 (en) * | 2018-10-25 | 2020-04-30 | 广州小鹏汽车科技有限公司 | Vehicle collision detection method and vehicle control system |
| WO2020187259A1 (en) * | 2019-03-18 | 2020-09-24 | 长城汽车股份有限公司 | Safety monitoring method and system for autonomous vehicle, and motion control system |
| WO2020187254A1 (en) * | 2019-03-18 | 2020-09-24 | 长城汽车股份有限公司 | Longitudinal control method and system for automatic driving vehicle |
| CN111746555A (en) * | 2019-03-26 | 2020-10-09 | 通用汽车环球科技运作有限责任公司 | Identification and avoidance of collision behavior |
| CN110435647A (en) * | 2019-07-26 | 2019-11-12 | 大连理工大学 | A kind of vehicle safety anticollision control method of the TTC based on rolling optimization parameter |
| CN111976638A (en) * | 2020-08-28 | 2020-11-24 | 广州市网优优信息技术开发有限公司 | Vehicle collision detection method and system based on vehicle-mounted intelligent terminal |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114633743A (en) | 2022-06-17 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104732805B (en) | A kind of dynamic early-warning method of automobile anti-rear end collision | |
| CN106043279B (en) | The lane shift control system and its control method influenced based on crosswind | |
| CN105109490B (en) | Method for judging sharp turn of vehicle based on three-axis acceleration sensor | |
| CN104742845B (en) | Dynamic early warning system for automobile rear-end collision preventing | |
| CN108466616B (en) | Method for automatically identifying collision event, storage medium and vehicle-mounted terminal | |
| CN105761323A (en) | Method and device for recognizing collision event based on on-board data | |
| CN104590275A (en) | Driving behavior analyzing method | |
| CN111649955A (en) | Performance evaluation method for vehicle-road cooperative automatic emergency braking system | |
| CN113702067B (en) | Self-adaptive cruise system evaluation system and method for commercial vehicle | |
| CN109774473B (en) | Speed limit control method based on camera and navigation data fusion | |
| CN106627590A (en) | Braking distance calculation method and device | |
| CN113640017A (en) | Test evaluation system and method for automatic emergency braking system of commercial vehicle | |
| CN111435160A (en) | In-loop testing method and system for vehicle radar device | |
| CN108062856B (en) | Vehicle collision detection system and method based on vehicle-mounted OBD interface | |
| CN105799709A (en) | Method and device for recognizing sudden turn of vehicle | |
| CN112046455B (en) | Automatic emergency braking method based on vehicle quality identification | |
| CN111976638B (en) | Vehicle collision detection method and system based on vehicle-mounted intelligent terminal | |
| Derbel | Driving style assessment based on the GPS data and fuzzy inference systems | |
| CN113968231B (en) | Intelligent driver model parameter determination method conforming to driver habits | |
| CN115158304A (en) | Automatic emergency braking control system and method | |
| CN114633743B (en) | Automatic driving vehicle and collision accident detection method and system thereof | |
| CN106494398A (en) | The reminding method and mobile terminal that prevent vehicle rear-end collision based on mobile terminal | |
| CN108956156A (en) | The performance test methods and its device of the seized system of vehicle | |
| Haspalamutgıl et al. | Adaptive switching method for adaptive cruise control | |
| CN111483458B (en) | Power system control method and device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| CB02 | Change of applicant information |
Country or region after: China Address after: No. 6, Yutong Road, Guancheng Hui District, Zhengzhou, Henan 450061 Applicant after: Yutong Bus Co.,Ltd. Address before: No.1, Shibali Heyu Road, Guancheng Hui District, Zhengzhou City, Henan Province Applicant before: ZHENGZHOU YUTONG BUS Co.,Ltd. Country or region before: China |
|
| CB02 | Change of applicant information | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |