CN115140029A

CN115140029A - Safety capability detection method and device for automatic driving automobile

Info

Publication number: CN115140029A
Application number: CN202210822767.7A
Authority: CN
Inventors: 黄朝胜; 张伟伟; 李骏; 黄毅
Original assignee: Tsinghua University
Current assignee: Tsinghua University
Priority date: 2022-07-12
Filing date: 2022-07-12
Publication date: 2022-10-04

Abstract

A safety capability detection method and device for an autonomous vehicle, the method comprising: when the automatic driving automobile and a first vehicle have collision risks, obtaining response control force output by the automatic driving automobile, wherein the first vehicle is a front vehicle of which the current lane driven by the automatic driving automobile is closest to the automatic driving automobile; determining a motion variation of the autonomous vehicle, and a relative speed and a relative distance of the autonomous vehicle and the first vehicle at a first time when the autonomous vehicle responds to braking; generating a double margin spectrum by using a preset double margin spectrum model according to the determined motion change parameters, the relative speed and the relative distance, wherein the double margin spectrum consists of a time margin spectrum and a control margin spectrum; determining safety capability data for the autonomous vehicle based on the generated double margin spectrum.

Description

Method and device for detecting safety capability of automatic driving automobile

Technical Field

The present disclosure relates to the field of automatic driving technologies, and in particular, to a method and an apparatus for detecting safety capability of an automatic driving vehicle.

Background

The safety evaluation of the automatic driving automobile is a key link of iterative development and feedback verification of the intelligent automobile, and the potential safety hazard and the capability defect of the intelligent automobile are difficult to find due to insufficient safety evaluation, so that the unexpected collision accident is easy to cause. In some technologies, the autopilot safety capability is only based on whether a collision is detected or not as a verification criterion, and lacks deep understanding and analysis of the autopilot system response and lacks an accurate observation model for differentiation and inspiration of the autopilot system response.

Disclosure of Invention

The application provides a safety capability detection method and device for an automatic driving automobile, and the method solves the problems that the automatic driving safety capability verification process is single, and the reasonableness and the flexibility are not high.

The application provides a safety capability detection method of an automatic driving automobile, which comprises the following steps:

when the automatic driving automobile and a first vehicle have collision risks, obtaining response control force output by the automatic driving automobile, wherein the first vehicle is a front vehicle of which the current lane driven by the automatic driving automobile is closest to the automatic driving automobile;

determining a motion variation parameter of the autonomous vehicle, and a relative speed and a relative distance of the autonomous vehicle and the first vehicle at a first time when the autonomous vehicle responds to braking;

generating a double margin spectrum by using a preset double margin spectrum model according to the determined motion change parameters, the relative speed and the relative distance, wherein the double margin spectrum consists of a time margin spectrum and a control margin spectrum;

determining safety capability data for the autonomous vehicle based on the generated double margin spectrum.

In an exemplary embodiment, the time margin spectrum is a spectrum with scene urgency as abscissa and time margin as ordinate;

the control margin spectrum is a spectrum with scene urgency as an abscissa and an operation margin as an ordinate.

In an exemplary embodiment, determining the change in motion parameter of the autonomous vehicle at the first time that the autonomous vehicle responds to braking comprises:

and at the first moment when the automatic driving automobile makes a braking response, acquiring the speed, the position and the actual response time of the automatic driving automobile at the first moment by using the automatic driving automobile control platform.

In an exemplary embodiment, the generating a double margin spectrum using a preset double margin spectrum model according to the determined motion variation parameter, the relative speed and the relative distance includes:

calculating a time response margin of the autonomous vehicle using a response time margin model;

calculating the braking margin of the automatic driving automobile by using the control margin model;

and generating a double margin spectrum according to the obtained response margin spectrum curve and the obtained control margin spectrum curve at the time level.

In one exemplary embodiment, the response time margin model is:

Tm＝(Sd-Sa)/V；

the response time margin model is the time difference Tm between the first moment Td and the preset theoretical latest response moment Ta;

and the Td is the braking response time of the automatic driving automobile, the Sd is the relative distance between the automatic driving automobile and the first vehicle corresponding to the first time, the Sa is the relative distance between the automatic driving automobile and the first vehicle corresponding to the theoretical latest response time, and the V is the running speed of the automatic driving automobile at the first time.

In an exemplary embodiment, the relative distance Sa of the autonomous vehicle to the first vehicle at the theoretical latest response time Ta is the distance required for the autonomous vehicle to brake with maximum brake intensity and for the relative speed to also just decrease to 0; the Sa is determined by:

Sa＝(V-V0)*(V-V0)/2/a0；

and V is the running speed of the automatic driving automobile at the first moment, V0 is the speed of the first automobile at the first moment, and a0 is the maximum braking strength of the automobile on the current road surface.

In an exemplary embodiment, the steering margin model: am = a0-a;

wherein a is response control force output by the decision of the automatic driving automobile at the first moment Td, a0 is the maximum braking strength of the automobile on the current road surface, and am is the braking margin of the collision avoidance capacity;

if am is greater than 0, representing that the control margin has a margin;

if am =0, then full force braking is represented;

if am <0, the maximum braking strength of the automobile on the current road surface is smaller than the expected collision avoidance braking strength a applied by automatic driving, and collision is inevitable under the working condition.

In an exemplary embodiment, the scene urgency is: c =1/TTC, TTC = Sd/(V-V0);

or C =1/HeadWay, headWay = Sd/V;

where Sd is a relative distance between the autonomous driving vehicle and the first vehicle at the first time, V is a speed of the autonomous driving vehicle at the first time, and V0 is a speed of the first vehicle at the first time.

In an exemplary embodiment, the determining safety capability data for the autonomous vehicle based on the generated double margin spectrum includes:

according to the generated double margin spectrum, when the collision avoidance control margin value is determined to be zero, an interval on the continuous scale of the response capability of the automatic driving automobile is obtained;

calculating the total area of the corresponding response time margin in the interval;

wherein the total area represents an accumulated evaluation of the safety capability of the autonomous vehicle on a continuous scale.

The application also provides a safety capability detection device of the automatic driving automobile, the memory is used for storing a program for detecting the safety capability of the automatic driving automobile, the processor is used for reading and executing the program for detecting the safety capability of the automatic driving automobile, and the method in any one of the embodiments is executed.

Compared with the related art, the application provides a safety capability detection method and a safety capability detection device for an automatic driving automobile, wherein the method comprises the following steps: when the automatic driving automobile and a first vehicle have collision risks, obtaining response control force output by the automatic driving automobile, wherein the first vehicle is a front vehicle of which the current lane driven by the automatic driving automobile is closest to the automatic driving automobile; determining a motion variation parameter of the autonomous vehicle, and a relative speed and a relative distance of the autonomous vehicle and the first vehicle at a first time when the autonomous vehicle responds to braking; generating a double margin spectrum by using a preset double margin spectrum model according to the determined motion change parameters, the relative speed and the relative distance, wherein the double margin spectrum consists of a time margin spectrum and a control margin spectrum; determining safety capability data for the autonomous vehicle based on the generated double margin spectrum. By the technical scheme, the problems of single verification process of the automatic driving safety capability, lack of rationality and low flexibility are solved.

Additional features and advantages of the application will be set forth in the description which follows, and in part will be obvious from the description, or may be learned by the practice of the application. Other advantages of the present application can be realized and attained by the instrumentalities and combinations particularly pointed out in the specification and the drawings.

Drawings

The accompanying drawings are included to provide an understanding of the present disclosure and are incorporated in and constitute a part of this specification, illustrate embodiments of the disclosure and together with the examples serve to explain the principles of the disclosure and not to limit the disclosure.

FIG. 1 is a flow chart of a method for detecting the safety capability of an autonomous vehicle according to an embodiment of the present application;

FIG. 2 is a schematic diagram of a safety capability detection apparatus for an autonomous vehicle according to an embodiment of the present application;

FIG. 3 is a block diagram of a method for automated driving safety capability evaluation based on stochastic responses in some exemplary embodiments;

FIG. 4 is a flow diagram of a method for automated driving safety capability assessment based on stochastic responses in some exemplary embodiments;

FIG. 5 is a schematic view of a safety assessment test scenario in some exemplary embodiments;

FIG. 6 is a simulation of operating conditions for safety evaluation in some exemplary embodiments;

FIG. 7 is a diagram of security evaluation process time nodes in some example embodiments;

FIG. 8 is a flow chart of a method for evaluating the safety capabilities of an autonomous vehicle in some exemplary embodiments;

FIG. 9 is a schematic illustration of a single response time line of an autonomous vehicle in some exemplary embodiments;

FIG. 10 is an example of the distribution of actual lines of response in multiple random scenarios in some example embodiments;

FIG. 11 is a depiction of point pairs of time/steering margins in a coordinate system for physical significance in some exemplary embodiments;

FIG. 12 is a graphical illustration of a double margin response model of an autonomous vehicle and its safety bandwidth values in some exemplary embodiments.

Detailed Description

The present application describes embodiments, but the description is illustrative rather than limiting and it will be apparent to those of ordinary skill in the art that many more embodiments and implementations are possible within the scope of the embodiments described herein. Although many possible combinations of features are shown in the drawings and discussed in the detailed description, many other combinations of the disclosed features are possible. Any feature or element of any embodiment may be used in combination with or instead of any other feature or element in any other embodiment, unless expressly limited otherwise.

The present application includes and contemplates combinations of features and elements known to those of ordinary skill in the art. The embodiments, features and elements disclosed in this application may also be combined with any conventional features or elements to form a unique inventive concept as defined by the claims. Any feature or element of any embodiment may also be combined with features or elements from other inventive aspects to form yet another unique inventive aspect, as defined by the claims. Thus, it should be understood that any of the features shown and/or discussed in this application may be implemented alone or in any suitable combination. Accordingly, the embodiments are not to be restricted except in light of the attached claims and their equivalents. Furthermore, various modifications and changes may be made within the scope of the appended claims.

Further, in describing representative embodiments, the specification may have presented the method and/or process as a particular sequence of steps. However, to the extent that the method or process does not rely on the particular order of steps set forth herein, the method or process should not be limited to the particular sequence of steps described. Other orders of steps are possible as will be understood by those of ordinary skill in the art. Therefore, the particular order of the steps set forth in the specification should not be construed as limitations on the claims. Further, the claims directed to the method and/or process should not be limited to the performance of their steps in the order written, and one skilled in the art can readily appreciate that the sequences may be varied and still remain within the spirit and scope of the embodiments of the present application.

The embodiment of the disclosure provides a safety capability detection method of an automatic driving automobile, as shown in fig. 1, the method includes steps S100-S130, which are specifically as follows:

s100, when the automatic driving automobile and a first vehicle have a collision risk, obtaining response control force output by the automatic driving automobile, wherein the first vehicle is a front vehicle of which the current lane where the automatic driving automobile runs is closest to the automatic driving automobile;

s110, determining a motion change parameter of the automatic driving automobile and the relative speed and the relative distance between the automatic driving automobile and the first vehicle at a first moment when the automatic driving automobile makes a braking response;

s120, generating a double margin spectrum by using a preset double margin spectrum model according to the determined motion change parameter, the relative speed and the relative distance, wherein the double margin spectrum consists of a time margin spectrum and a control margin spectrum;

and S130, determining safety capacity data of the automatic driving automobile according to the generated double margin spectrum.

In the present embodiment, in most automatic driving scenes, the collision is a safety bottom line of an automatic driving automobile, and in the present embodiment, it is assumed that the process is a complex nonlinear system, initially in a safe state, after a certain moment of movement, a danger is sensed, and in a safe operable space, two limit safe braking forms exist: firstly, when the line is marked at the earliest response, the line immediately responds, performs deceleration braking with small braking force, and then enters a new safety state relative to the front vehicle; and secondly, when the line is marked at the latest response time, the brake is instantly responded, full-force braking is carried out by the maximum braking force, and then the vehicle enters a new safety state relative to the front vehicle.

Physically, as shown in fig. 6, the earliest response time line is based on the real-time speed/position of the current computing time, and as long as the collision risk can be sensed, the earliest response time line can be considered as the earliest response time line; and the reticle is actually calculated by physical quantities such as relative distance, relative speed and the like when the vehicle responds at the latest, and is related to the maximum collision avoidance braking capacity a0 of the vehicle.

In this embodiment, the first vehicle is a vehicle in front of the autonomous vehicle in a current lane where the autonomous vehicle travels, the vehicle being closest to the autonomous vehicle, and the speed of the first vehicle is less than the running speed of the autonomous vehicle.

The response control force of the decision output of the automatic driving automobile is output from a control platform of the automatic driving automobile, and the response control force is determined according to the current vehicle operation parameters by the control platform of the automatic driving automobile.

In an exemplary embodiment, the amount of time between the actual response time line and the latest response time line of the autonomous vehicle is defined as a response time margin, representing a graceful response capability. Under different working conditions, the actual response time scales of the automatic driving automobile are different, and at the moment, response time margin normalization is carried out. The time margin spectrum is a spectrum with the scene urgency as an abscissa and the time margin as an ordinate.

If the expected control capability at the actual response time of the autonomous vehicle is a, the collision avoidance control margin is defined as a ₀ -a. Wherein, if the collision avoidance control margin is positive, a redundant safety capability of the free edge is embodied; if the tolerance value is zero, the brake is emergent full force brake, and the tension safety capability is slightly displayed; if the margin value is negative, the actual collision avoidance capability of the automobile is smaller than the expected applied collision avoidance control capability, and the marking line moves to the left side of the marking line at the latest response time at the moment, so that the collision is unavoidable. The control margin spectrum is a spectrum with scene urgency as an abscissa and an operation margin as an ordinate.

In an exemplary embodiment, the offline representation of the multidimensional safety margin can be obtained by combining the physical definitions of the response time margin and the collision avoidance control capability margin, then the reciprocal of Headway or the reciprocal of TTC (predicted collision time) of the autonomous vehicle and the front obstacle can be used for representing the degree of urgency of the working condition by combining a numerical interpolation fitting method, and the two margins are respectively used as vertical coordinates to form the evaluation result of the random response time of the safety capability of the autonomous vehicle. Wherein, the scene urgency is: c =1/TTC, TTC = Sd/(V-V0); or C =1/HeadWay, headWay = Sd/V; where Sd is the relative distance between the autonomous driving vehicle and the first vehicle at the first time, V is the speed of the autonomous driving vehicle at the first time, and V0 is the speed of the first vehicle at the first time.

In an exemplary embodiment, said determining a motion variation of the autonomous vehicle at a first instant in time when the autonomous vehicle responds to braking comprises: and at the first moment when the automatic driving automobile makes a braking response, acquiring the speed, the position and the actual response time of the automatic driving automobile at the first moment by using the automatic driving automobile control platform.

In an exemplary embodiment, the generating a double margin spectrum using a preset double margin spectrum model according to the determined motion variation parameter, the relative speed and the relative distance includes: calculating a time response margin of the autonomous vehicle using a response time margin model; calculating the braking margin of the automatic driving automobile by using an operation margin model; and generating a double margin spectrum according to the obtained response margin spectrum curve and the obtained control margin spectrum curve at the time level. In the present embodiment, for a scene parameter, a point pair of the response margin and the manipulation margin is obtained, such as a single-response time line diagram of the autonomous vehicle shown in fig. 9, where the first line counted from the left side is the latest response line; the third line counted from the left side is the earliest response line; the second line from the left is a line for actual response time of a certain time; in one scenario, a pair of response and steering margins is obtained. The schematic of the physical meaning of the point pair for each time margin/steering margin in the coordinate system is shown in fig. 11, and the abscissa represents the scene urgency as: c =1/TTC or C =1/HeadWay. The part of the ordinate axis above the abscissa axis represents the operation time margin corresponding to the response, and the part of the ordinate axis below the abscissa axis represents the collision avoidance capacity margin corresponding to the response. The first point on the left side in the figure represents, and under the working condition, the AD responds to a better time margin and a collision avoidance capacity margin; under the same type of working conditions, the response capability of the left second point pair representation is obviously inferior to that of the first point pair AD system. The third point on the left in the figure represents that at the last moment, with maximum braking force, a collision can just be avoided. The fourth pair on the left in the figure represents that even with maximum braking force, a collision still occurs; however, if no collision occurs, an additional virtual braking force (a negative margin) is required. The desired braking force is large. The fifth point pair on the left side in the figure represents that the working condition is more urgent and the collision is more likely to happen considering that the more the abscissa is rightward; even if the situation is beyond the inherent capability of the vehicle, the emergency degree may cause the collision to be difficult to avoid. In the test under the state, the collision avoidance capacity margin is necessarily negative, but the time margin is not necessarily negative.

As shown in fig. 10, the distribution of the actual response lines in a plurality of random scenes, that is, in a plurality of scenes, a point pair of a response margin and a manipulation margin is obtained for each scene, and a plurality of point pairs of a response margin and a manipulation margin are obtained for a plurality of scenes; and finally, forming a response margin spectrum curve and a control margin spectrum curve at a time level according to the plurality of pairs of the response margin and the control margin, and generating a double margin spectrum according to the formed response margin spectrum curve and the formed control margin spectrum curve.

In one exemplary embodiment, the response time margin model is:

Tm＝(Sd-Sa)/V；

and the moment Td is the moment when the automatic driving automobile makes a braking response, sd is the relative distance between the automatic driving automobile and the first vehicle corresponding to the first moment, sa is the relative distance between the automatic driving automobile and the first vehicle corresponding to the theoretical latest response moment, and V is the running speed of the automatic driving automobile at the first moment. In the present embodiment, the relative distance Sa of the autonomous vehicle to the first vehicle at the theoretical latest response time Ta is the distance required for the autonomous vehicle to brake with the maximum brake strength and for the relative speed to also decrease to exactly 0; the Sa is determined by:

Sa＝(V-V0)*(V-V0)/2/a0；

In an exemplary embodiment, the steering margin model: am = a0-a;

if am is greater than 0, representing that the control margin has allowance;

if am =0, then full force braking is represented;

if am <0, the maximum braking intensity of the automobile on the current road surface is smaller than the expected collision avoidance braking intensity a applied by automatic driving, and collision is inevitable under the working condition.

An embodiment of the present disclosure further provides a traffic signal control device, as shown in fig. 2, the device includes: a memory 210 and a processor 220; the memory 210 is used for storing a program for detecting the safety capability of the automatic driving automobile, and the processor 220 is used for reading and executing the program for detecting the safety capability of the automatic driving automobile and executing the method of any one of the above embodiments.

Example one

FIG. 3 shows a structural block diagram of an automatic driving safety capability evaluation method based on random response according to the embodiment, wherein the evaluation method comprises a natural driving full-scene domain model, a traffic scene random excitation model, an automatic driving automobile control platform, a response equivalent double margin spectrum model and continuous scale safety capability evaluation;

the natural driving full scene domain model is used for acquiring scene parameters of a target virtual scene;

the traffic scene random excitation model is used for acquiring a multi-dimensional motion parameter of a target virtual scene vehicle;

the automatic driving automobile control platform is used for feeding back multiple motion change parameters such as the speed, the acceleration and the like of an automatic driving automobile;

the response equivalent double-margin spectrum model is used for generating a response time margin and a collision avoidance control margin by adopting a variation parameter of an automatic driving automobile control platform.

And determining the accumulated evaluation value of the safety capacity of the automatic driving automobile under the continuous scale according to the response time margin and the collision avoidance control margin generated by the response equivalent double-margin spectrum model so as to evaluate the safety capacity of the automatic driving automobile.

In the present example, as shown in fig. 4, the natural driving full scene domain model includes multi-type scene screening, scene vehicle positioning detection model, vehicle multi-dimensional motion parameter extraction, motion parameter probability distribution modeling, and natural driving scene complexity description.

The traffic scene random excitation model comprises perception, evaluation and decision of an automatic driving automobile in different scenes, and is combined with automobile system dynamics to form external motion scene excitation and an impact response model.

The automatic driving automobile control platform comprises an automatic driving automobile speed position and light rain fog adjustment, a virtual instrument measurement and control platform, a VTEHIL experiment platform, an automatic driving elastic safety envelope and an automatic driving model with adjustable parameters.

The response equivalent double-margin spectrum model comprises a response time margin, a collision avoidance control margin and a working condition urgency degree, and a double expression model of the time response margin and the collision avoidance control margin of an actual response time point is constructed; and combining continuous excitation of different scenes and the multidimensional margin offline representation of numerical interpolation fitting to obtain a safety capability evaluation index system of the automatic driving automobile under continuous scale.

Example two

The following describes a safety capability detection method process of an automatic driving automobile by using an example:

the safety capacity of this autopilot's safety capacity detection's test scenario, this safety capacity detection scenario schematic diagram as shown in fig. 5, including experimental control platform, virtual operating mode platform, autopilot and automobile wheel hub unsettled platform. The test control platform is in communication connection with the virtual working condition platform, the test control platform is in communication connection with the automatic driving automobile, the automatic driving automobile is placed on the automobile hub suspension platform, and the automatic driving automobile identifies scenes in the virtual working condition platform through a sensor of the automatic driving automobile.

And secondly, setting a working condition simulation diagram for detecting the safety capability of the automatic driving automobile, as shown in fig. 6, considering various scenes such as random deceleration of vehicles in the same front lane, switching in/out of vehicles in adjacent lanes or ramp lanes, transverse lane switching of the vehicle and the like, wherein the simulation diagram shows multi-parameter expression forms such as a collision-free latest response time scale line, a collision-free earliest response time scale line, an actual response time scale line, a response time margin, a collision avoidance control margin, a maximum acceleration, an actual response time acceleration, an elastic domain constrained by an automatic driving decision parameter and the like.

Thirdly, setting a time node graph for detecting the safety capability of the automatic driving automobile, as shown in fig. 7, wherein the abscissa represents time, the ordinate is provided with three modules, namely an experiment control platform, a virtual working condition platform and the automatic driving automobile, a diamond point on each module represents the response of the module at the time, and an arrow represents a transmission path of response parameters, for example: the first arrow on the left side in the figure indicates that the test control platform A sends the selected working condition and the set parameters such as weather, illumination, road conditions and the like to the virtual working condition platform B at the first time point; the second arrow on the left side of the figure indicates that the test control platform a sends the set autonomous vehicle speed V to the autonomous vehicle C at a second point in time; the third arrow on the left side in the figure indicates that the automatic driving automobile C detects that the distance from the target automobile C is L meters at the third time point, the parameter is sent to the test control platform A, the test control platform A responds after receiving the parameter, and the parameter for setting the target automobile speed and starting to decelerate at the fourth time point is sent to the virtual working condition platform B; feeding back the speed and the acceleration of the actual response moment to the test control platform A by the automatic driving automobile C at a fifth time point to calculate TTC; and feeding back a response time margin spectrum and a control margin to the test control platform A by the automatic driving automobile C at a sixth time point, and calculating continuous initial safety capability evaluation at the sixth time point. Fig. 7 can visually and clearly illustrate the response points and the response sequences of the modules in the test scenario.

Fourthly, establishing and training a double margin spectrum model;

in this step, a double margin spectrum is established, which consists of a time margin spectrum and a manipulation margin spectrum; the time margin spectrum is a spectrum with scene urgency as an abscissa and time margin as an ordinate;

As shown in fig. 8, the distribution of the actual response lines under multiple scenarios is selected to train the dual margin spectrum model, and the training process may include the following steps:

step 1, selecting a virtual working condition platform (cut-in/cut-out and deceleration) and scene weather conditions, illumination intensity, road surface conditions and the like by a test control platform;

step 2, the test control platform sets the speed of the automatic driving automobile on the automobile hub suspension platform to be V = N km/h (N =20, 30, 40, 50, 60, 70 and 80);

step 3, starting providing a speed V for the target vehicle in the virtual working condition and starting decelerating by the test control platform at a position L = Mm (M =10, 20, 30, 40, 50, 60) away from the target vehicle on the lane;

step 4, the virtual working condition platform and the automatic driving automobile feed back multiple motion parameters such as actual response time speed, acceleration and the like to the test control platform, and the test control platform calculates TTC (predicted time to collide);

step 5, measuring a response time margin and a collision avoidance control margin at the actual response moment of the automatic driving automobile, feeding the response time margin and the collision avoidance control margin back to the test control platform, and constructing a dual expression model of the response time margin and the collision avoidance control margin at the response moment;

and 6, repeatedly executing the steps 1 to 5 to obtain a dual margin expression model of the continuous scale response time margin and the collision avoidance control margin of the automatic driving automobile. As shown in fig. 12, when the margin value of the collision avoidance control is zero, it represents that the autonomous vehicle has just reached the vehicle capability boundary (zero crossing), the cutoff emergency (bandwidth) representing the emergency degree of the emergency condition can be handled, the total area of the response time margin corresponding to the bandwidth represents the sum of the margins of a response capability, and the area of a positive value of the margin value of the collision avoidance control also represents a safety capability, and the area of the dual margin expression model is calculated, thereby completing the numerical calculation and evaluation of the safety capability of the autonomous vehicle in a continuous scale.

In the embodiment, the safety capability detection method of the automatic driving automobile can synchronously form a double margin spectrum model related to time margin-control capability margin in a double coordinate system through response observation of a plurality of random traffic scenes by responding to instant braking response of the automatic driving automobile under different traffic scenes, defining a time margin model and a control margin model in the automobile control dimension, and defining a capability bandwidth model in the models so as to carry out quantitative index evaluation on the automatic driving safety capability on a continuous scale; the problems that the process of verifying the safety capability of the automatic driving is single, and the reasonability and the flexibility are poor are solved. The method integrates interdisciplines such as vehicle system dynamics/kinematics, classical control theory and the like, considers the whole scene area of the automatic driving automobile, inputs variables according to multiple dimensions such as relative distance, relative speed and the like, and judges the safety attribute and the automatic driving safety capability of the automatic driving automobile in a certain variable space by combining the response of the input variables to the space.

It will be understood by those of ordinary skill in the art that all or some of the steps of the methods, systems, functional modules/units in the devices disclosed above may be implemented as software, firmware, hardware, or suitable combinations thereof. In a hardware implementation, the division between functional modules/units mentioned in the above description does not necessarily correspond to the division of physical components; for example, one physical component may have multiple functions, or one function or step may be performed by several physical components in cooperation. Some or all of the components may be implemented as software executed by a processor, such as a digital signal processor or microprocessor, or as hardware, or as an integrated circuit, such as an application specific integrated circuit. Such software may be distributed on computer readable media, which may include computer storage media (or non-transitory media) and communication media (or transitory media). The term computer storage media includes volatile and nonvolatile, removable and non-removable media implemented in any method or technology for storage of information such as computer readable instructions, data structures, program modules or other data, as is well known to those of ordinary skill in the art. Computer storage media includes, but is not limited to, RAM, ROM, EEPROM, flash memory or other memory technology, CD-ROM, digital Versatile Disks (DVD) or other optical disk storage, magnetic cassettes, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other medium which can be used to store the desired information and which can accessed by a computer. In addition, communication media typically embodies computer readable instructions, data structures, program modules or other data in a modulated data signal such as a carrier wave or other transport mechanism and includes any information delivery media as known to those skilled in the art.

Claims

1. A method for detecting a safety capability of an autonomous vehicle, the method comprising:

2. The method of detecting a safety capability of an autonomous vehicle according to claim 1,

the time margin spectrum is a spectrum with scene urgency as an abscissa and time margin as an ordinate;

3. The method of claim 2, wherein determining the motion variation of the autonomous vehicle at the first time of the braking response of the autonomous vehicle comprises:

and at the first moment when the automatic driving automobile makes braking response, acquiring the speed, the position and the actual response time of the automatic driving automobile at the first moment by using the automatic driving automobile control platform.

4. The method of claim 3, wherein the generating a double margin spectrum using a preset double margin spectrum model according to the determined motion variation, relative speed and relative distance comprises:

5. The method of detecting a safety capability of an autonomous vehicle according to claim 4,

the response time margin model is:

Tm＝(Sd-Sa)/V；

6. The method of detecting a safety capability of an autonomous vehicle according to claim 5,

the relative distance Sa between the autonomous vehicle and the first vehicle corresponding to the theoretical latest response time Ta is a distance required for the autonomous vehicle to brake with the maximum brake strength and for the relative speed to be just reduced to 0; the Sa is determined by:

Sa＝(V-V0)*(V-V0)/2/a0；

7. The method of detecting a safety capability of an autonomous vehicle according to claim 6,

the control margin model is as follows: am = a0-a;

if am is greater than 0, representing that the control margin has a margin;

if am =0, then full force braking is represented;

8. The method of detecting a safety capability of an autonomous vehicle according to claim 7,

the scene urgency is as follows: c =1/TTC, TTC = Sd/(V-V0);

or C =1/HeadWay, headWay = Sd/V;

where Sd is the relative distance between the autonomous driving vehicle and the first vehicle at the first time, V is the speed of the autonomous driving vehicle at the first time, and V0 is the speed of the first vehicle at the first time.

9. The method of detecting a safety capability of an autonomous vehicle according to claim 1,

determining safety capability data for the autonomous vehicle based on the generated double margin spectrum, comprising:

10. A safety capability detecting apparatus for an autonomous vehicle, the apparatus comprising: a memory and a processor; wherein the memory is used for storing a program for performing safety capability detection of an autonomous vehicle, and the processor is used for reading and executing the program for performing safety capability detection of an autonomous vehicle and executing the method of any one of claims 1-9.