CN114973656A - Method, device, equipment, medium and product for evaluating traffic interaction performance - Google Patents

Method, device, equipment, medium and product for evaluating traffic interaction performance Download PDF

Info

Publication number
CN114973656A
CN114973656A CN202210507919.4A CN202210507919A CN114973656A CN 114973656 A CN114973656 A CN 114973656A CN 202210507919 A CN202210507919 A CN 202210507919A CN 114973656 A CN114973656 A CN 114973656A
Authority
CN
China
Prior art keywords
traffic
determining
target
action
traffic state
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN202210507919.4A
Other languages
Chinese (zh)
Inventor
周谷越
袁基睿
龚江涛
杨凡
王鲲
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Tsinghua University
Apollo Zhilian Beijing Technology Co Ltd
Original Assignee
Tsinghua University
Apollo Zhilian Beijing Technology Co Ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Tsinghua University, Apollo Zhilian Beijing Technology Co Ltd filed Critical Tsinghua University
Priority to CN202210507919.4A priority Critical patent/CN114973656A/en
Publication of CN114973656A publication Critical patent/CN114973656A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G08—SIGNALLING
    • G08G—TRAFFIC CONTROL SYSTEMS
    • G08G1/00—Traffic control systems for road vehicles
    • G08G1/01—Detecting movement of traffic to be counted or controlled
    • G08G1/0104—Measuring and analyzing of parameters relative to traffic conditions
    • G08G1/0125—Traffic data processing

Landscapes

  • Chemical & Material Sciences (AREA)
  • Analytical Chemistry (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Traffic Control Systems (AREA)

Abstract

The disclosure provides a method, a device, equipment, a medium and a product for evaluating traffic interaction performance, and relates to the technical field of artificial intelligence, in particular to the technical field of intelligent traffic and automatic driving. The specific implementation scheme comprises the following steps: determining test traffic states corresponding to N moments based on cross-reference information aiming at traffic participating objects in the target test scene according to the initial traffic state corresponding to the target test scene, wherein N is an integer greater than 1; determining index quantitative evaluation values corresponding to the N moments based on preset performance indexes according to the initial traffic state and the test traffic state; and determining a performance evaluation value aiming at the target test scene according to the index quantitative evaluation values corresponding to the N moments, wherein the traffic state comprises the traffic state value and the state value distribution probability of the traffic participators in the target test scene.

Description

Method, device, equipment, medium and product for evaluating traffic interaction performance
Technical Field
The disclosure relates to the technical field of artificial intelligence, in particular to the technical field of intelligent traffic and automatic driving, and can be applied to scenes such as evaluation of traffic interaction performance.
Background
The evaluation of the traffic interaction performance is of great significance for ensuring the safe driving of vehicles and improving the traffic efficiency. However, in some scenarios, the evaluation of the traffic interaction performance has the phenomena of low evaluation efficiency and poor evaluation effect.
Disclosure of Invention
The present disclosure provides a method, apparatus, device, medium, and product for evaluating traffic interaction performance.
According to an aspect of the present disclosure, there is provided a method for evaluating traffic interaction performance, including: determining test traffic states corresponding to N moments according to an initial traffic state corresponding to a target test scene and based on cross-reference information aiming at traffic participating objects in the target test scene, wherein N is an integer greater than 2; according to the initial traffic state and the test traffic state, determining index quantitative evaluation values corresponding to the N moments based on preset performance indexes; and determining a performance evaluation value aiming at the target test scene according to the index quantitative evaluation values corresponding to the N moments, wherein the traffic state comprises a traffic state value and a state value distribution probability of traffic participating objects in the target test scene.
According to another aspect of the present disclosure, there is provided an apparatus for evaluating traffic interaction performance, including: the system comprises a first processing module, a second processing module and a third processing module, wherein the first processing module is used for determining test traffic states corresponding to N moments according to an initial traffic state corresponding to a target test scene and based on cross reference information aiming at traffic participating objects in the target test scene, and N is an integer greater than 2; the second processing module is used for determining an index quantitative evaluation value corresponding to the N moments and based on a preset performance index according to the initial traffic state and the test traffic state; and a third processing module, configured to determine a performance evaluation value for the target test scenario according to the index quantitative evaluation values corresponding to the N moments, where a traffic state includes a traffic state value and a state value distribution probability of a traffic participating object in the target test scenario.
According to another aspect of the present disclosure, there is provided an electronic device including: at least one processor and a memory communicatively coupled to the at least one processor. Wherein the memory stores instructions executable by the at least one processor, the instructions being executable by the at least one processor to enable the at least one processor to perform the above-described method of assessing traffic interaction performance.
According to another aspect of the present disclosure, there is provided a non-transitory computer-readable storage medium storing computer instructions for causing the computer to perform the above-described method of evaluating traffic interaction performance.
According to another aspect of the present disclosure, a computer program product is provided, comprising a computer program which, when executed by a processor, implements the above-described method of assessing traffic interaction performance.
It should be understood that the statements in this section do not necessarily identify key or critical features of the embodiments of the present disclosure, nor do they limit the scope of the present disclosure. Other features of the present disclosure will become apparent from the following description.
Drawings
The drawings are included to provide a better understanding of the present solution and are not to be construed as limiting the present disclosure. Wherein:
fig. 1 schematically illustrates a system architecture of a traffic interaction performance evaluation method and apparatus according to an embodiment of the present disclosure;
FIG. 2 schematically illustrates a flow chart of a method of assessing traffic interaction performance according to an embodiment of the present disclosure;
FIG. 3 schematically illustrates a flow chart of a method of assessing traffic interaction performance according to another embodiment of the present disclosure;
FIG. 4 schematically illustrates a schematic diagram of an evaluation process of traffic interaction performance according to an embodiment of the present disclosure;
FIG. 5 schematically illustrates a schematic diagram of a target test scenario according to an embodiment of the present disclosure;
fig. 6 schematically shows a block diagram of an evaluation device of traffic interaction performance according to an embodiment of the present disclosure;
fig. 7 schematically illustrates a block diagram of an electronic device for evaluation of traffic interaction performance in accordance with an embodiment of the present disclosure.
Detailed Description
Exemplary embodiments of the present disclosure are described below with reference to the accompanying drawings, in which various details of embodiments of the present disclosure are included to assist understanding, and which are to be considered as merely exemplary. Accordingly, those of ordinary skill in the art will recognize that various changes and modifications of the embodiments described herein can be made without departing from the scope and spirit of the present disclosure. Also, descriptions of well-known functions and constructions are omitted in the following description for clarity and conciseness.
The terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the disclosure. The terms "comprises," "comprising," and the like, as used herein, specify the presence of stated features, steps, operations, and/or components, but do not preclude the presence or addition of one or more other features, steps, operations, or components.
All terms (including technical and scientific terms) used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs, unless otherwise defined. It is noted that the terms used herein should be interpreted as having a meaning that is consistent with the context of this specification and should not be interpreted in an idealized or overly formal sense.
Where a convention analogous to "at least one of A, B and C, etc." is used, in general such a construction is intended in the sense one having skill in the art would understand the convention (e.g., "a system having at least one of A, B and C" would include but not be limited to systems that have a alone, B alone, C alone, a and B together, a and C together, B and C together, and/or A, B, C together, etc.).
The embodiment of the disclosure provides a traffic interaction performance evaluation method. The method of the embodiment comprises the following steps: determining test traffic states corresponding to N moments according to an initial traffic state corresponding to a target test scene and based on cross-reference information for traffic-participating objects in the target test scene, wherein N is an integer greater than 2, determining index quantitative evaluation values corresponding to the N moments and based on preset performance indexes according to the initial traffic state and the test traffic states, and determining performance evaluation values for the target test scene according to the index quantitative evaluation values corresponding to the N moments, wherein the traffic states comprise traffic state values and state value distribution probabilities of the traffic-participating objects in the target test scene.
Fig. 1 schematically shows a system architecture of a traffic interaction performance evaluation method and apparatus according to an embodiment of the present disclosure. It should be noted that fig. 1 is only an example of a system architecture to which the embodiments of the present disclosure may be applied to help those skilled in the art understand the technical content of the present disclosure, and does not mean that the embodiments of the present disclosure may not be applied to other devices, systems, environments or scenarios.
The system architecture 100 according to this embodiment may include a requesting terminal 101, a network 102, and a server 103. Network 102 is the medium used to provide communication links between requesting terminals 101 and server 103. Network 102 may include various connection types, such as wired, wireless communication links, or fiber optic cables, to name a few. The server 103 may be an independent physical server, a server cluster or a distributed system formed by a plurality of physical servers, or a cloud server providing basic cloud computing services such as cloud services, cloud computing, network services, middleware services, and the like.
The requesting terminal 101 interacts with the server 103 through the network 102 to receive or transmit data or the like. The request terminal 101 is configured to initiate an evaluation request of traffic interaction performance to the server 103, for example, and the request terminal 101 is further configured to provide an initial traffic state corresponding to the target test scenario to the server 103, where the initial traffic state includes a traffic state value and a state value distribution probability of a traffic participant in the target test scenario based on an initial time.
The server 103 may be a server providing various services, for example, a background processing server (for example only) performing traffic interaction performance evaluation according to an initial traffic state corresponding to the target test scenario provided by the requesting terminal 101.
For example, the server 103 determines a test traffic state corresponding to N times, N being an integer greater than 2, based on the cross-reference information for the traffic-participating objects in the target test scene from the initial traffic state corresponding to the target test scene provided by the requesting terminal 101, determines an index quantitative evaluation value based on a preset performance index corresponding to the N times from the initial traffic state and the test traffic state, and determines a performance evaluation value for the target test scene from the index quantitative evaluation values corresponding to the N times, the traffic state including a traffic state value and a state value distribution probability of the traffic-participating objects in the target test scene.
It should be noted that the method for evaluating traffic interaction performance provided by the embodiment of the present disclosure may be executed by the server 103. Accordingly, the device for evaluating traffic interaction performance provided by the embodiment of the present disclosure may be disposed in the server 103. The method for evaluating traffic interaction performance provided by the embodiment of the present disclosure may also be executed by a server or a server cluster that is different from the server 103 and is capable of communicating with the requesting terminal 101 and/or the server 103. Accordingly, the evaluation device for traffic interaction performance provided by the embodiment of the present disclosure may also be disposed in a server or a server cluster that is different from the server 103 and is capable of communicating with the requesting terminal 101 and/or the server 103.
It should be understood that the number of requesting terminals, networks, and servers in fig. 1 is merely illustrative. There may be any number of requesting terminals, networks, and servers, as desired for an implementation.
The embodiment of the present disclosure provides an evaluation method of traffic interaction performance, and the following describes the evaluation method of traffic interaction performance according to an exemplary embodiment of the present disclosure with reference to fig. 2 to 4 in conjunction with the system architecture of fig. 1. The method for evaluating traffic interaction performance of the embodiment of the present disclosure may be performed by the server 103 shown in fig. 1, for example.
Fig. 2 schematically illustrates a flow chart of a method of evaluating traffic interaction performance according to an embodiment of the present disclosure.
As shown in fig. 2, the method 200 for evaluating traffic interaction performance according to the embodiment of the present disclosure may include operations S210 to S230, for example.
In operation S210, test traffic states corresponding to N times are determined based on the mutual reference information for the traffic participant in the target test scenario according to the initial traffic state corresponding to the target test scenario, where N is an integer greater than 2.
In operation S220, index quantitative evaluation values based on preset performance indexes corresponding to N times are determined according to the initial traffic state and the test traffic state.
In operation S230, a performance evaluation value for the target test scenario is determined according to the index quantitative evaluation values corresponding to the N times, and the traffic state includes a traffic state value and a state value distribution probability of the traffic participant in the target test scenario.
The following illustrates exemplary operation flows of the traffic interaction performance evaluation method according to the present embodiment.
Exemplarily, according to an initial traffic state corresponding to the target test scenario, based on the cross-reference information for the traffic participant in the target test scenario, the test traffic states corresponding to N times are determined, where N is an integer greater than 2.
The goal testing scenarios may include, for example, interactive gambling-like scenarios, fault or extreme scenarios, environmental security-like scenarios, and so forth. The interactive game-like scenes may include, for example, an over-the-horizon following scene (over-the-horizon, i.e., beyond-the-range of viewing distances), a lane change conflict scene, an unprotected turn scene, etc. The fault or extreme scenario may include, for example, an abnormal obstacle scenario, a vehicle fault scenario, a road blockage/blockage scenario, etc. The environmental security scene may include, for example, a road construction scene, a rain, snow, fog and weather scene, and the like.
The initial traffic state may include, for example, traffic state values and state value distribution probabilities based on the initial time of day for the traffic participant in the target test scenario. Traffic-engaging objects may include, for example, traffic signals, automobiles, non-automobiles, pedestrians, short obstacles, and the like, and automobiles may include, for example, autonomous vehicles and non-autonomous vehicles.
The traffic state values in the target test scenario may include, for example, a following vehicle speed, a following vehicle headway, an intersection collision vehicle speed, an intersection collision distance, and the like. The state value distribution probability in the target test scenario may, for example, satisfy a known probability distribution, which may be derived from statistical data, for example. For example, the following vehicle speed may be subjected to a log-normal distribution, the following vehicle head time distance may be subjected to a negative exponential distribution, the intersection conflict vehicle speed may be subjected to a poisson distribution, and the reciprocal of the intersection conflict distance may be subjected to the poisson distribution.
The initial traffic state may indicate a scene type of the target test scene, and the initial traffic state may be represented by, for example, a hyper-parameter and a hyper-parameter probability distribution. For example, the hyperparameter { v } may be utilized 1 、v 2 H represents the traffic state value in the over-the-horizon following scenario, v 1 For example, the current vehicle speed, v 2 For example, the vehicle speed of the preceding vehicle, and h may be, for example, the headway between the current vehicle and the preceding vehicle. v. of 1 、v 2 For example, a lognormal distribution may be satisfied, and h, for example, a negative exponential distribution may be satisfied.
Exemplarily, in v 1 For the purpose of illustration, v can be represented by formula (1) 1 Distribution q (v) of 1 )，
Figure BDA0003636801920000061
Mu and sigma are basic parameters of lognormal distribution, mu represents mean, sigma represents standard deviation, and v 1 And the mu and the sigma can be obtained by fitting real traffic data. Suppose v 1 Satisfying a velocity range v min ，v max ]Q (v) can be represented by the formula (2) 1 ) Carrying out normalization calculation to obtain v 1 Probability distribution p (v) 1 )，
Figure BDA0003636801920000062
The probability distribution p (h) of h can be calculated by equation (3),
Figure BDA0003636801920000063
Figure BDA0003636801920000064
Figure BDA0003636801920000065
the average headway is represented and can be obtained according to real traffic data statistics. For example,
Figure BDA0003636801920000066
v 1 、v 2 h is an independent variable, and a traffic state value { v) can be calculated by the formula (4) 1 、v 2 H } probability distribution.
P(v 1 、v 2 、h)＝p(v 1 )p(v 2 )p(h) (4)
The traffic participant in the target test scene can interact based on the initial traffic state and the interactive reference information to obtain test traffic states corresponding to the N moments. For example, a test traffic state corresponding to the 1 st time may be determined based on the cross-reference information and the initial traffic state. And determining the test traffic state corresponding to the Nth moment based on the mutual reference information and the test traffic state corresponding to the (N-1) th moment. N2, … … N, the test traffic status may include traffic status values and status value distribution probabilities of the road participants based on the test time.
The cross-reference information may include fused reference information from at least one reference source. The at least one reference source may comprise at least one of: vehicle-mounted terminal, roadside terminal and high in the clouds. The fused reference information may include at least one of: fusion perception information, fusion decision information and fusion control information.
The vehicle-road cloud integrated automatic driving model is taken as an example for explanation, and the vehicle-road cloud integrated automatic driving model can perform multi-level cooperative sensing, cooperative decision and cooperative control from dimensions such as a vehicle-mounted terminal, a road side terminal, a cloud end and the like, so that a complete automatic driving closed loop is realized. The vehicle-mounted terminal, the road side terminal and the cloud end in the vehicle-road-cloud integrated automatic driving model can play roles in layers such as complementation, redundancy and reinforcement.
The complementary effect is taken as an example for explanation, aiming at the aspect of cooperative sensing, as the vehicle-mounted terminal possibly has the defects of limited sensing range, high possibility of being shielded, high influence of environment and light, limited sensing precision and the like, the roadside terminal and the cloud end can fully exert the advantages of wide sensing range, long-time continuous observation, easiness in engineering and the like, the cooperative sensing is performed with vehicle-end sensing, the cooperative sensing under the types of environments such as shielding, beyond visual range, dynamic and static blind areas and the like is realized, and the sensing capability of the vehicle-mounted terminal can be effectively improved. Aiming at the aspects of cooperative decision and cooperative control, under the condition of meeting an extreme traffic environment or a complex driving scene, the automatic driving vehicle is easy to have the phenomena of emergency braking or forced take-over and the like, and the automatic driving vehicle can be re-planned with a path through cooperative decision or cooperative control with a road side terminal and a cloud end, or a cloud end automatic driving system directly controls the vehicle to help the vehicle safely pass through the extreme traffic environment or the complex driving scene.
The redundant function is taken as an example for explanation, and for the aspect of cooperative sensing, under the condition that the sensing function of the vehicle-mounted terminal normally operates, the cooperative sensing information provided by the road side terminal or the cloud terminal can play a redundant role. The redundant cooperative sensing information is beneficial to improving the environmental sensing precision and ensuring the safe driving of the automatic driving vehicle.
Taking the enhancement as an example, for the aspect of traffic signal recognition, the vehicle-mounted terminal needs to recognize and locate the traffic signal in a three-dimensional space through an optical sensor, and to predict the change of the light color of the traffic signal. Because of the possibility of glare, interference of tail lights of vehicles, stroboflash of an LED (Light-Emitting Diode), standard-reaching delay of lamp brightness, damage/aging of lamps, displacement of movable signal lamps, shaking of suspended signal lamps, multiphase matching, beyond visual range, blind areas of visual fields, shielding of dynamic and static obstacles, perception interference factors such as abnormal weather and the like, accurate and reliable semantic information of the signal lamps can be obtained through low-code-rate coding through butt joint information of a road side terminal and a low-complexity signal lamp machine, intention information such as dynamic countdown is obtained through beyond visual range, and cooperative perception information provided by the road side terminal or a cloud can play a role in strengthening.
In response to obtaining the test traffic states corresponding to the N moments, quantitative evaluation values of the indexes corresponding to the N moments based on the preset performance indexes may be determined according to the initial traffic state and the test traffic state. And determining a performance evaluation value aiming at the target test scene according to the index quantitative evaluation values corresponding to the N moments. The preset performance index may include, for example, a collision probability index, a traffic efficiency index, a somatosensory comfort index, and the like.
Through the embodiment of the disclosure, the test traffic states corresponding to N moments are determined based on the cross-reference information for the traffic participant in the target test scene according to the initial traffic state corresponding to the target test scene, wherein N is an integer greater than 2, the index quantitative evaluation value based on the preset performance index corresponding to the N moments is determined according to the initial traffic state and the test traffic state, and the performance evaluation value for the target test scene is determined according to the index quantitative evaluation value corresponding to the N moments. By determining the test traffic states corresponding to the N moments, the traffic performance evaluation precision aiming at the target test scene can be effectively improved, and the traffic interaction performance based on the interactive reference information can be intuitively quantized. According to the initial traffic state corresponding to the target test scene, the performance evaluation value aiming at the target test scene is determined, so that the evaluation efficiency of the traffic interaction performance can be effectively improved, the evaluation cost of the traffic interaction performance is effectively reduced, and the safe and efficient running of the automatic driving vehicle is ensured.
Fig. 3 schematically shows a flow chart of a method of evaluating traffic interaction performance according to another embodiment of the present disclosure.
As shown in fig. 3, the method 300 for evaluating traffic interaction performance according to the embodiment of the present disclosure may include operations S310 to S350 and operations S220 to S230, for example.
In operation S310, a target interaction action to be performed by the traffic participant is determined using the traffic interaction model corresponding to the target test scenario.
In operation S320, an action attribute and a time distribution probability of the traffic participant performing the target interactive action are determined based on the initial traffic state and the interactive reference information.
In operation S330, test traffic states corresponding to the N moments are determined according to the motion attributes and the time distribution probabilities of the traffic participant performing the target interactive motion.
In operation S220, index quantitative evaluation values based on preset performance indexes corresponding to N times are determined according to the initial traffic state and the test traffic state.
In operation S230, a performance evaluation value for the target test scenario is determined according to the index quantitative evaluation values corresponding to the N times, and the traffic state includes a traffic state value and a state value distribution probability of the traffic participant in the target test scenario.
The following illustrates exemplary operation flows of the traffic interaction performance evaluation method according to the present embodiment.
Illustratively, a traffic interaction model corresponding to the target test scene is utilized to determine a target interaction action to be executed by the traffic participant. The traffic interaction model corresponding to the target test scenario may include, for example, an over-the-horizon following model, a lane change conflict model, an unprotected steering model, a road blocking/blocking model, and the like.
The traffic interaction model may include, for example, a rendering module for simulating a target test scenario and for visualizing traffic interaction events in the target test scenario, a fusion perception module, a fusion decision module, a fusion control module, and the like. The fusion perception module is used for fusing perception information from at least one perception source, and the at least one perception source can comprise an in-vehicle terminal, a roadside terminal and a cloud terminal. The fusion decision module is used for fusing decision information from at least one decision source, and the at least one decision source can comprise an on-board terminal, a road side terminal and a cloud terminal. The fusion control module is used for fusing control information from at least one control source, and the at least one control source may include an in-vehicle terminal, a roadside terminal and a cloud terminal, for example. The fusion control module can be used for controlling the traffic participation objects in the target test scene to interact.
And determining the target interaction action to be executed by the traffic participant by using the traffic interaction model matched with the target test scene. The beyond-the-horizon following scene is taken as an example for explanation, and the target interaction action to be executed by the current vehicle comprises sensing the deceleration of the previous vehicle and completing the corresponding deceleration action. Taking a lane change conflict scene as an example for explanation, the target interaction action to be executed by the current vehicle comprises sensing that the previous vehicle changes lanes and finishing a corresponding deceleration action. Taking a pedestrian ghost probe scene as an example for explanation, the target interaction action to be executed by the current vehicle includes sensing that a pedestrian crosses a lane and completes a corresponding deceleration action.
And determining action attributes and time distribution probability of the traffic participant executing the target interactive action based on the initial traffic state and the interactive reference information. And determining the test traffic states corresponding to the N moments according to the action attributes and the time distribution probability of the traffic participant executing the target interactive action.
The target interactive action is described as a deceleration action as an example, and the action attribute of the deceleration action may include information such as a vehicle position, a vehicle speed, and a vehicle acceleration when the deceleration action is performed, for example. And forming the time distribution probability of the target interactive action based on the action occurrence probability of executing the target interactive action at each moment in the N moments.
By way of example, the action attribute and the action occurrence probability of the traffic participant executing the target interactive action based on the 1 st moment can be determined based on the interactive reference information and the initial traffic state. And determining action attributes and action occurrence probability of the traffic participation object executing the target interactive action based on the nth moment based on the interactive reference information and the test traffic state corresponding to the (n-1) th moment. And N is 2, … … N, and the traffic participant executes the action occurrence probability of the target interaction action based on the N moments to form the time distribution probability of the target interaction action.
And determining a test traffic state corresponding to the 1 st moment according to the initial traffic state and the action attribute and the action occurrence probability of the traffic participant executing the target interactive action based on the 1 st moment. And determining the test traffic state corresponding to the nth moment according to the test traffic state corresponding to the (N-1) th moment and the action attribute and the action occurrence probability of the traffic participant executing the target interactive action based on the nth moment.
And obtaining action attributes and action occurrence probabilities of the traffic participant executing the target interactive action based on the current time based on the interactive reference information aiming at the traffic participant according to the test traffic state value and the test state value distribution probability corresponding to the previous time, and further obtaining the test traffic state corresponding to the current time. The traffic interaction performance evaluation accuracy aiming at the target test scene can be effectively improved, the evaluation efficiency of the traffic interaction performance is favorably improved, and the safety and the ordered interaction of traffic participation objects in the real traffic scene are favorably ensured.
The cross-reference information includes fused reference information from at least one reference source. The at least one reference source comprises at least one of: the system comprises a vehicle-mounted terminal, a road side terminal and a cloud. The fused reference information includes at least one of: fusion perception information, fusion decision information and fusion control information.
And determining action attributes and action occurrence probability of the traffic participation object executing the target interactive action based on the nth moment based on the interactive reference information and the test traffic state corresponding to the (n-1) th moment. Taking the beyond-the-horizon following scene as an example for explanation, the test traffic state P (v) corresponding to the (n-1) th moment can be based on the cross-reference information 1 、v 2 H), determining the action attribute and the action occurrence probability of the current vehicle executing the deceleration action at the nth moment. For example, the determination of the traffic participation object based on the nth time can also be executed based on the mutual reference information and the test traffic state corresponding to at least part of the first n-1 timesAction attributes and action occurrence probabilities of the target interaction actions.
And the traffic participant executes the action occurrence probability of the target interactive action based on the N moments to form the time distribution probability of the target interactive action. The uncertainty in the time of occurrence of the action of the target interaction may be caused by, for example, a perceptual uncertainty. Perceptual uncertainty may include, for example, uncertainty arising from hardware detection, model identification, multi-sensor fusion, and the like. For example, in an over-the-horizon following scenario, the perceived uncertainty may be a function of distance.
For example, the traffic state value corresponding to the nth time may be determined according to the test traffic state corresponding to the (N-1) th time and the action attribute of the traffic participant performing the target interaction based on the nth time. And determining the state value distribution probability corresponding to the nth moment according to the action occurrence probability of the traffic participant executing the target interactive action based on the nth moment.
In response to obtaining the test traffic states corresponding to the N moments, quantitative evaluation values of the indexes corresponding to the N moments based on the preset performance indexes may be determined according to the initial traffic state and the test traffic state. For example, the index quantitative evaluation value based on the preset performance index for the nth time may be determined according to the initial traffic state and the test traffic state corresponding to the nth time. An index quantitative evaluation value based on a preset performance index for the nth time can also be determined according to the initial traffic state and the test traffic state corresponding to at least part of the previous n times.
In one example, the predetermined performance metric includes a collision probability metric. The predicted positions of conflicts between traffic participants may be determined based on the initial traffic state and the scene type of the target test scene. And determining the probability of collision of the traffic participating objects based on the collision prediction positions according to the traffic state values and the state value distribution probabilities corresponding to the N moments, and quantifying the evaluation value as an index.
And the positions where the collision is possible among the traffic participating objects form the collision prediction position. The position of a possible collision between traffic participating objects can be determined according to the initial traffic state and the scene type of the target test scene. The probability of collision of the traffic participant based on each moment can be determined according to the test traffic states corresponding to the N moments. For example, it is possible to determine, from the test traffic states corresponding to N times, a time difference at which the traffic-participating object arrives at the collision prediction position based on the times, and determine, based on the time difference, a probability of collision of the traffic-participating object based on the times, as an index quantization evaluation value corresponding to the times.
The traffic interaction performance of the target test scene is evaluated by determining the index quantitative evaluation value based on the collision probability index corresponding to the N moments, so that the interaction safety performance of the target test scene can be efficiently evaluated, the evaluation precision of the traffic interaction performance can be effectively improved, the evaluation efficiency of the traffic interaction performance can be effectively improved, and the safe and efficient operation of traffic participants in a real traffic scene can be guaranteed.
In another example, the predetermined performance metric includes a traffic efficiency metric. The area to be evaluated in the target test scene can be determined according to the initial traffic state and the scene type of the target test scene. And determining the predicted time length required by the traffic participant to pass through the area to be evaluated according to the traffic state values and the state value distribution probabilities corresponding to the N moments, so as to be used as an index quantitative evaluation value.
The traffic interaction performance of the target test scene is evaluated by determining the traffic efficiency index-based quantitative evaluation value corresponding to the N moments, so that the high-efficiency evaluation of the traffic capacity of the target test scene is facilitated, the evaluation efficiency of the traffic interaction performance can be effectively improved, and the evaluation precision of the traffic interaction performance can be effectively improved.
For example, according to the initial traffic state and the scene type of the target test scene, at least one of the following regions in the target test scene may be determined as the region to be evaluated: an interactive gaming-like area, a fault/extreme scene area, and an environmental security-like area.
The interactive game type area can be a danger area caused by conflict, and can comprise pedestrian ghost probes, vehicle lane change conflict, unprotected vehicle steering, multi-lane merging, no-signal lamp intersection and other types of areas. The fault/extreme scene areas may include, for example, vehicle trapped type areas such as vehicle faults, unusual obstacles, extreme congestion, road congestion/blockages, and the like. The environmental safety-type areas may include, for example, road construction, traffic safety accidents, rainy, snowy, foggy weather, surface water/ice, road surface collapse, tunnel fires, and the like.
And determining the predicted time length required by the traffic participation object to pass through the area to be evaluated based on each test traffic state according to the test traffic states corresponding to the N moments, and taking the predicted time length as an index quantitative evaluation value corresponding to each moment. For example, the predicted time length required for the traffic participant to pass through the area to be evaluated based on the corresponding traffic state value may be determined as an index quantitative evaluation value based on the traffic efficiency index corresponding to each time, according to the traffic state value and the state value distribution probability corresponding to N times. And determining a performance evaluation value for the target test scenario according to the index quantitative evaluation values corresponding to the N moments.
For example, in the case where the number of target test scenarios is more than one, the performance evaluation value for the automatic driving model may be determined from the performance evaluation values corresponding to a plurality of target test scenarios. For example, the performance evaluation value E for the automatic driving model can be calculated by equation (5),
Figure BDA0003636801920000121
s represents a test scene set, and the test scene set can comprise a plurality of target test scenes. For any target test scenario, θ 0 An initial traffic state value, P (theta), representing a traffic participant in a target test scenario 0 ) Representing the initial state value distribution probability. Theta T A test traffic state value, P (theta), representing traffic participants in the target test scenario T ) Denotes the test state value distribution probability, T1. E (theta) T ) Is shown for eachAnd testing the index quantitative evaluation value of the traffic state, namely representing the index quantitative evaluation value corresponding to each moment.
By the embodiment of the disclosure, the target interactive action to be executed by the traffic participant is determined, and the action attribute and the time distribution probability of the traffic participant executing the target interactive action are determined based on the initial traffic state and the interactive reference information. And determining the test traffic states corresponding to the N moments according to the action attribute and the time distribution probability of the traffic participant executing the target interaction action. And determining a performance evaluation value for the target test scenario according to the initial traffic state and the test traffic states corresponding to the N moments. The traffic interaction performance evaluation method has the advantages that the traffic interaction performance evaluation precision aiming at the target test scene can be effectively improved, the traffic interaction performance evaluation efficiency can be effectively improved, the safety and the ordered interaction of traffic participation objects in the real traffic scene can be guaranteed, and the safety driving of automatic driving vehicles can be guaranteed.
Fig. 4 schematically shows a schematic diagram of an evaluation process of traffic interaction performance according to an embodiment of the present disclosure.
As shown in fig. 4, in the process of evaluating the traffic interaction performance, a scene distribution model 401 is used to determine an initial traffic state corresponding to a target test scene according to any target test scene in the sampling simulation set S. The initial traffic state includes an initial traffic state value theta of the traffic participant in the target test scenario 0 And initial state value distribution probability P (theta) 0 )。
The traffic interaction model 402 is utilized to determine test traffic states corresponding to the N moments based on the cross-reference information for the traffic participants in the target test scenario according to the initial traffic state corresponding to the target test scenario. The test traffic state includes a test traffic state value theta of the traffic participant T And test state value distribution probability P (theta) T ) N, N being an integer greater than 2.
The traffic interaction model 402 may include, for example, a rendering module, a fusion awareness module, a fusion decision module, and a fusion control module. The rendering module is used, for example, to simulate a target test scenario and to visualize traffic interaction events in the target test scenario. The fusion perception module is used for fusing perception information from at least one reference source, the fusion decision module is used for fusing decision information from at least one reference source, the fusion control module is used for fusing control information from at least one reference source, and the fusion control module can be used for controlling traffic participation objects in a target test scene to interact.
The cross-reference information may include fused reference information from at least one reference source. The at least one reference source may for example comprise at least one of: vehicle-mounted terminal, roadside terminal and high in the clouds. The fused reference information may include, for example, at least one of: fusion perception information, fusion decision information and fusion control information.
Using the traffic interaction model 402, based on the initial traffic state value θ 0 And cross reference information, determining the test traffic state value theta corresponding to the 1 st moment 1 And a test state value distribution probability P (theta) 1 ). Based on the test traffic state value theta corresponding to the (n-1) th time n-1 And cross reference information, determining the test traffic state value theta corresponding to the Nth moment n And a test state value distribution probability P (theta) n )，n＝2,......N。
Using the evaluation model 403, index quantitative evaluation values based on preset performance indexes corresponding to N times are determined according to the initial traffic state and the test traffic state, and a performance evaluation value for the target test scenario is determined according to the index quantitative evaluation values corresponding to N times.
Illustratively, the probability P (θ) is distributed according to the initial state values using the evaluation model 403 0 ) And a test state value distribution probability P (theta) T ) Determining index quantitative evaluation values E (theta) corresponding to N moments based on preset performance indexes T ) N is an integer greater than 2. Quantizing the evaluation value E (theta) according to the index T ) And determining a performance evaluation value for the target test scene.
And determining the test traffic states corresponding to the N moments according to the initial traffic state corresponding to the target test scene. And determining a performance evaluation value aiming at the target test scene according to the initial traffic state and the test traffic states corresponding to the N moments. The traffic interaction performance evaluation method has the advantages that the traffic interaction performance evaluation precision aiming at the target test scene can be effectively improved, the traffic interaction performance evaluation efficiency can be effectively improved, the traffic interaction performance evaluation cost can be effectively reduced, the safety and the ordered interaction of traffic participation objects in the real traffic scene can be guaranteed, and the safety driving of automatic driving vehicles can be guaranteed.
FIG. 5 schematically shows a schematic diagram of a target test scenario according to an embodiment of the present disclosure.
As shown in FIG. 5, the target test scenario may be, for example, a vehicle failure scenario. In a vehicle failure scenario to be tested, the target vehicle 501 may be an autonomous vehicle, with the failed vehicle 502 located in the direction of travel lane of the target vehicle 501 and the oncoming vehicle 503 located in a lane adjacent to the direction of travel lane of the target vehicle 501.
In performing traffic performance evaluation based on a vehicle failure scenario, an initial traffic state corresponding to the vehicle failure scenario may be set. The initial traffic state may include, for example, a traffic state value and a state value distribution probability based on the initial time of the target vehicle 501, the faulty vehicle 502, and the oncoming vehicle 503, respectively. The traffic state values may include, for example, information of vehicle position, vehicle speed, headway, and the like.
And driving the target vehicle 501, the fault vehicle 502 and the opposite vehicle 503 to interact by using a traffic interaction model corresponding to a vehicle fault scene, so as to obtain test traffic states corresponding to N moments. The test traffic state may include, for example, a traffic state value and a state value distribution probability of the target vehicle 501 and the opposing vehicle 503 based on each test time, respectively. The traffic interaction model may be, for example, a vehicle road cloud integrated simulation model.
And determining index quantitative evaluation values corresponding to the N moments based on the preset performance indexes according to the initial traffic state and the test traffic state. And determining a performance evaluation value for the vehicle fault scene according to the index quantitative evaluation values corresponding to the N moments. The preset performance indicators may include, for example, traffic efficiency indicators, collision probability indicators, and the like.
From the performance evaluation value for the vehicle breakdown scenario, the target vehicle 501 can identify the running state of the breakdown vehicle 502 based on the fusion perception information. The fusion perception information may include vehicle-side perception information, road-side perception information, and cloud perception information, for example. The road side perception information and the cloud side perception information have the advantages of wide perception range and long continuous observation time, can perform perception complementation and reinforcement with the vehicle side perception information, and can assist the target vehicle 501 in identifying the running state of the fault vehicle 502 as fault parking.
The target vehicle 501 performs the detour around the faulty vehicle 502 after recognizing that the running state of the faulty vehicle 502 is the faulty stop. The oncoming vehicle 503 may obtain a lane change detour driving decision of the target vehicle 501 through the lane-cloud integrated simulation model. The oncoming vehicle 503 can perform a deceleration action based on the lane-change detour driving decision of the target vehicle 501, thereby effectively reducing the occurrence probability of the collision event. The vehicle road cloud integrated simulation model can provide driving behavior information increment of the target vehicle 501 for the oncoming vehicle 503, the global vehicle cooperative decision making capability can provide information increment such as driving intention for the automatic driving vehicle, the single-lane opposite traffic capability can be effectively improved, and feasible and effective driving behavior decision making can be adopted in a complex traffic scene.
Fig. 6 schematically shows a block diagram of an evaluation device of traffic interaction performance according to an embodiment of the present disclosure.
As shown in fig. 6, the apparatus 600 for evaluating traffic interaction performance according to the embodiment of the present disclosure includes, for example, a first processing module 610, a second processing module 620, and a third processing module 630.
A first processing module 610, configured to determine, according to an initial traffic state corresponding to a target test scenario, test traffic states corresponding to N times based on cross-reference information for traffic participants in the target test scenario, where N is an integer greater than 2; the second processing module 620 is configured to determine, according to the initial traffic state and the test traffic state, index quantitative evaluation values corresponding to N moments based on preset performance indexes; and a third processing module 630, configured to determine a performance evaluation value for the target test scenario according to the index quantitative evaluation values corresponding to the N times, where the traffic state includes a traffic state value and a state value distribution probability of a traffic participant in the target test scenario.
Through the embodiment of the disclosure, the test traffic states corresponding to N moments are determined based on the cross-reference information for the traffic participant in the target test scene according to the initial traffic state corresponding to the target test scene, wherein N is an integer greater than 2, the index quantitative evaluation value based on the preset performance index corresponding to the N moments is determined according to the initial traffic state and the test traffic state, and the performance evaluation value for the target test scene is determined according to the index quantitative evaluation value corresponding to the N moments. By determining the test traffic states corresponding to the N moments, the traffic performance evaluation precision aiming at the target test scene can be effectively improved, and the traffic interaction performance based on the interactive reference information can be intuitively quantized. According to the initial traffic state corresponding to the target test scene, the performance evaluation value aiming at the target test scene is determined, so that the evaluation efficiency of the traffic interaction performance can be effectively improved, the evaluation cost of the traffic interaction performance is effectively reduced, and the safe and efficient running of the automatic driving vehicle is ensured.
According to an embodiment of the present disclosure, a first processing module includes: the first processing submodule is used for determining a target interaction action to be executed by a traffic participant by using a traffic interaction model corresponding to a target test scene; the second processing submodule is used for determining the action attribute and the time distribution probability of the traffic participant executing the target interactive action based on the initial traffic state and the interactive reference information; and the third processing submodule is used for determining the test traffic state corresponding to the N moments according to the action attribute and the time distribution probability of the traffic participant executing the target interaction action.
According to an embodiment of the present disclosure, the second processing submodule includes: the first processing unit is used for determining the action attribute and the action occurrence probability of the traffic participation object executing the target interactive action based on the 1 st moment based on the interactive reference information and the initial traffic state; and a second processing unit, configured to determine, based on the cross-reference information and the test traffic state corresponding to the (N-1) th time, an action attribute and an action occurrence probability of the traffic participant performing the target interaction action based on the nth time, where N is 2, … … N, and the action occurrence probability of the traffic participant performing the target interaction action based on the N times constitutes a time distribution probability.
According to an embodiment of the present disclosure, the third processing submodule includes: the third processing unit is used for determining a test traffic state corresponding to the 1 st moment according to the initial traffic state and the action attribute and the action occurrence probability of the traffic participant executing the target interactive action based on the 1 st moment; and the fourth processing unit is used for determining the test traffic state corresponding to the nth moment according to the test traffic state corresponding to the (n-1) th moment and the action attribute and the action occurrence probability of the traffic participant executing the target interactive action based on the nth moment.
According to an embodiment of the present disclosure, the cross-reference information comprises fused reference information from at least one reference source; the at least one reference source comprises at least one of: the system comprises a vehicle-mounted terminal, a roadside terminal and a cloud terminal; and the fused reference information comprises at least one of: fusion perception information, fusion decision information and fusion control information.
According to an embodiment of the present disclosure, the preset performance index includes a collision probability index; a second processing module comprising: the fourth processing submodule is used for determining a conflict prediction position between traffic participation objects according to the initial traffic state and the scene type of the target test scene; and a fifth processing sub-module for determining a probability of collision of the traffic participation object based on the collision prediction position as an index quantization evaluation value, based on the traffic state values and the state value distribution probabilities corresponding to the N times.
According to an embodiment of the present disclosure, the preset performance index includes a passage efficiency index; a second processing module comprising: the sixth processing sub-module is used for determining an area to be evaluated in the target test scene according to the initial traffic state and the scene type of the target test scene; and the seventh processing submodule is used for determining the predicted time length required by the traffic participation object to pass through the area to be evaluated according to the traffic state values and the state value distribution probability corresponding to the N moments, and the predicted time length is used as an index quantitative evaluation value.
It should be noted that in the technical solutions of the present disclosure, the processes of collecting, storing, using, processing, transmitting, providing, disclosing and the like of the related information are all in accordance with the regulations of the related laws and regulations, and do not violate the customs of the public order.
The present disclosure also provides an electronic device, a readable storage medium, and a computer program product according to embodiments of the present disclosure.
Fig. 7 schematically illustrates a block diagram of an electronic device for evaluation of traffic interaction performance in accordance with an embodiment of the present disclosure.
FIG. 7 illustrates a schematic block diagram of an example electronic device 700 that can be used to implement embodiments of the present disclosure. The electronic device 700 is intended to represent various forms of digital computers, such as laptops, desktops, workstations, personal digital assistants, servers, blade servers, mainframes, and other appropriate computers. The electronic device may also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions, are meant to be examples only, and are not meant to limit implementations of the disclosure described and/or claimed herein.
As shown in fig. 7, the device 700 comprises a computing unit 701, which may perform various suitable actions and processes according to a computer program stored in a Read Only Memory (ROM)702 or a computer program loaded from a storage unit 708 into a Random Access Memory (RAM) 703. In the RAM 703, various programs and data required for the operation of the device 700 can also be stored. The computing unit 701, the ROM 702, and the RAM 703 are connected to each other by a bus 704. An input/output (I/O) interface 705 is also connected to bus 704.
Various components in the device 700 are connected to the I/O interface 705, including: an input unit 706 such as a keyboard, a mouse, or the like; an output unit 707 such as various types of displays, speakers, and the like; a storage unit 708 such as a magnetic disk, optical disk, or the like; and a communication unit 709 such as a network card, modem, wireless communication transceiver, etc. The communication unit 709 allows the device 700 to exchange information/data with other devices via a computer network, such as the internet, and/or various telecommunication networks.
Computing unit 701 may be a variety of general and/or special purpose processing components with processing and computing capabilities. Some examples of the computing unit 701 include, but are not limited to, a Central Processing Unit (CPU), a Graphics Processing Unit (GPU), various dedicated Artificial Intelligence (AI) computing chips, various computing units running deep learning model algorithms, a Digital Signal Processor (DSP), and any suitable processor, controller, microcontroller, and so forth. The calculation unit 701 executes the respective methods and processes described above, such as the object description generation method. For example, in some embodiments, the object description generation method may be implemented as a computer software program tangibly embodied on a machine-readable medium, such as storage unit 708. In some embodiments, part or all of a computer program may be loaded onto and/or installed onto device 700 via ROM 702 and/or communications unit 709. When loaded into RAM 703 and executed by the computing unit 701, may perform one or more steps of the above described method of assessing traffic interaction performance. Alternatively, in other embodiments, the computing unit 701 may be configured in any other suitable way (e.g., by means of firmware) to implement the method of evaluating traffic interaction performance.
Various implementations of the systems and techniques described here above may be implemented in digital electronic circuitry, integrated circuitry, Field Programmable Gate Arrays (FPGAs), Application Specific Integrated Circuits (ASICs), Application Specific Standard Products (ASSPs), system on a chip (SOCs), Complex Programmable Logic Devices (CPLDs), computer hardware, firmware, software, and/or combinations thereof. These various embodiments may include: implemented in one or more computer programs that are executable and/or interpretable on a programmable system including at least one programmable processor, which may be special or general purpose, receiving data and instructions from, and transmitting data and instructions to, a storage system, at least one input device, and at least one output device.
Program code for implementing the methods of the present disclosure may be written in any combination of one or more programming languages. These program codes may be provided to a processor or controller of a general purpose computer, special purpose computer, or other programmable model training apparatus, such that the program codes, when executed by the processor or controller, cause the functions/acts specified in the flowchart and/or block diagram to be performed. The program code may execute entirely on the machine, partly on the machine, as a stand-alone software package partly on the machine and partly on a remote machine or entirely on the remote machine or server.
In the context of this disclosure, a machine-readable medium may be a tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device. The machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. More specific examples of a machine-readable storage medium would include an electrical connection based on one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.
To provide for interaction with an object, the systems and techniques described here can be implemented on a computer having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to an object; and a keyboard and a pointing device (e.g., a mouse or a trackball) by which objects can provide input to the computer. Other kinds of devices may also be used to provide for interaction with an object; for example, feedback provided to the subject can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the object can be received in any form, including acoustic, speech, or tactile input.
The systems and techniques described here can be implemented in a computing system that includes a back-end component (e.g., as a data server), or that includes a middleware component (e.g., an application server), or that includes a front-end component (e.g., an object computer having a graphical object interface or a web browser through which objects can interact with an implementation of the systems and techniques described here), or any combination of such back-end, middleware, or front-end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: local Area Networks (LANs), Wide Area Networks (WANs), and the Internet.
The computer system may include clients and servers. A client and server are generally remote from each other and typically interact through a communication network. The relationship of client and server arises by virtue of computer programs running on the respective computers and having a client-server relationship to each other. The server may be a cloud server, a server of a distributed system, or a server with a combined blockchain.
It should be understood that various forms of the flows shown above, reordering, adding or deleting steps, may be used. For example, the steps described in the present disclosure may be executed in parallel, sequentially or in different orders, and are not limited herein as long as the desired results of the technical solutions disclosed in the present disclosure can be achieved.
The above detailed description should not be construed as limiting the scope of the disclosure. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions may be made, depending on design requirements and other factors. Any modification, equivalent replacement, and improvement made within the spirit and principle of the present disclosure should be included in the scope of protection of the present disclosure.

Claims (17)

1. A method of assessing traffic interaction performance, comprising:
determining test traffic states corresponding to N moments according to an initial traffic state corresponding to a target test scene and based on cross-reference information aiming at traffic participating objects in the target test scene, wherein N is an integer greater than 2;
according to the initial traffic state and the test traffic state, determining index quantitative evaluation values corresponding to the N moments based on preset performance indexes; and
determining a performance evaluation value for the target test scenario from the index quantitative evaluation values corresponding to the N times,
the traffic state comprises a traffic state value and a state value distribution probability of a traffic participant in the target test scene.
2. The method of claim 1, wherein the determining test traffic states corresponding to N time instants based on cross-reference information for traffic-engaging objects in a target test scenario from an initial traffic state corresponding to the target test scenario comprises:
determining a target interaction action to be executed by the traffic participant by using a traffic interaction model corresponding to the target test scene;
determining action attributes and time distribution probabilities of the traffic participant executing the target interactive action based on the initial traffic state and the interactive reference information; and
and determining the test traffic state corresponding to the N moments according to the action attribute and the time distribution probability of the traffic participant executing the target interactive action.
3. The method of claim 2, wherein the determining an action attribute and a time-distributed probability that the traffic participant performs the target interaction action based on the initial traffic state and the cross-reference information comprises:
based on the interaction reference information and the initial traffic state, determining action attributes and action occurrence probability of the traffic participant executing the target interaction action based on the 1 st moment; and
determining action attributes and action occurrence probabilities of the traffic participation objects executing the target interactive action based on the nth time based on the interactive reference information and the test traffic state corresponding to the (n-1) th time,
wherein N is 2, … … N, and the action occurrence probability of the traffic participant performing the target interaction based on the N time instants constitutes the time distribution probability.
4. The method of claim 3, wherein the determining the test traffic state corresponding to the N moments according to the action attribute and the time distribution probability of the traffic participant performing the target interaction comprises:
determining a test traffic state corresponding to the 1 st moment according to the initial traffic state and the action attribute and action occurrence probability of the traffic participant executing the target interaction action based on the 1 st moment; and
and determining the test traffic state corresponding to the Nth moment according to the test traffic state corresponding to the (N-1) th moment and the action attribute and action occurrence probability of the traffic participant executing the target interaction action based on the nth moment.
5. The method of any one of claims 1 to 4,
the cross-reference information comprises fused reference information from at least one reference source;
the at least one reference source comprises at least one of: the system comprises a vehicle-mounted terminal, a roadside terminal and a cloud terminal; and
the fused reference information includes at least one of: fusion perception information, fusion decision information and fusion control information.
6. The method of claim 1, wherein the preset performance indicators comprise collision probability indicators; the determining of the index quantitative evaluation value corresponding to the N moments and based on the preset performance index according to the initial traffic state and the test traffic state comprises the following steps:
determining a conflict prediction position between the traffic participation objects according to the initial traffic state and the scene type of the target test scene; and
and determining the probability of the collision of the traffic participation object based on the collision prediction position according to the traffic state values and the state value distribution probability corresponding to the N moments, and taking the probability as the index quantitative evaluation value.
7. The method of claim 1, wherein the preset performance metric comprises a traffic efficiency metric; the determining of the index quantitative evaluation value corresponding to the N moments and based on the preset performance index according to the initial traffic state and the test traffic state comprises the following steps:
determining a region to be evaluated in the target test scene according to the initial traffic state and the scene type of the target test scene; and
and determining the predicted time length required by the traffic participant to pass through the area to be evaluated according to the traffic state values and the state value distribution probabilities corresponding to the N moments, and taking the predicted time length as the index quantitative evaluation value.
8. An evaluation device of traffic interaction performance, comprising:
the system comprises a first processing module, a second processing module and a third processing module, wherein the first processing module is used for determining test traffic states corresponding to N moments according to an initial traffic state corresponding to a target test scene and based on cross reference information aiming at traffic participating objects in the target test scene, and N is an integer greater than 2;
the second processing module is used for determining index quantitative evaluation values corresponding to the N moments and based on preset performance indexes according to the initial traffic state and the test traffic state; and
a third processing module for determining a performance evaluation value for the target test scenario according to the index quantitative evaluation values corresponding to the N moments,
the traffic state comprises a traffic state value and a state value distribution probability of a traffic participant in the target test scene.
9. The apparatus of claim 8, wherein the first processing module comprises:
the first processing submodule is used for determining a target interaction action to be executed by the traffic participant by using a traffic interaction model corresponding to the target test scene;
the second processing submodule is used for determining action attributes and time distribution probability of the traffic participant for executing the target interactive action based on the initial traffic state and the interactive reference information; and
and the third processing submodule is used for determining the test traffic state corresponding to the N moments according to the action attribute and the time distribution probability of the traffic participant executing the target interaction action.
10. The apparatus of claim 9, wherein the second processing submodule comprises:
the first processing unit is used for determining action attributes and action occurrence probabilities of the traffic participation objects executing the target interactive action based on the 1 st moment based on the interactive reference information and the initial traffic state; and
a second processing unit, configured to determine, based on the cross-reference information and a test traffic state corresponding to an nth-1 th time, an action attribute and an action occurrence probability of the traffic participant performing the target interaction action based on the nth time,
wherein N is 2, … … N, and the traffic participant constitutes the time distribution probability based on the action occurrence probability of the target interactive action executed at the N time instants.
11. The method apparatus of claim 10, wherein the third processing sub-module comprises:
the third processing unit is used for determining a test traffic state corresponding to the 1 st moment according to the initial traffic state and the action attribute and action occurrence probability of the traffic participant executing the target interactive action based on the 1 st moment; and
and the fourth processing unit is used for determining the test traffic state corresponding to the Nth moment according to the test traffic state corresponding to the (N-1) th moment and the action attribute and action occurrence probability of the traffic participant executing the target interactive action based on the nth moment.
12. The apparatus of any one of claims 8 to 11,
the cross-reference information comprises fused reference information from at least one reference source;
the at least one reference source comprises at least one of: the system comprises a vehicle-mounted terminal, a roadside terminal and a cloud terminal; and
the fused reference information includes at least one of: fusion perception information, fusion decision information and fusion control information.
13. The apparatus of claim 8, wherein the preset performance indicators comprise a collision probability indicator; the second processing module comprises:
the fourth processing submodule is used for determining a conflict prediction position between the traffic participation objects according to the initial traffic state and the scene type of the target test scene; and
and the fifth processing sub-module is used for determining the probability of collision of the traffic participation object based on the collision prediction position according to the traffic state values and the state value distribution probabilities corresponding to the N moments, and the probability is used as the index quantitative evaluation value.
14. The apparatus of claim 8, wherein the preset performance metric comprises a traffic efficiency metric; the second processing module comprises:
the sixth processing submodule is used for determining an area to be evaluated in the target test scene according to the initial traffic state and the scene type of the target test scene; and
and the seventh processing submodule is used for determining the predicted time length required by the traffic participation object to pass through the area to be evaluated according to the traffic state values and the state value distribution probabilities corresponding to the N moments, and the predicted time length is used as the index quantitative evaluation value.
15. An electronic device, comprising:
at least one processor; and
a memory communicatively coupled to the at least one processor; wherein the content of the first and second substances,
the memory stores instructions executable by the at least one processor to enable the at least one processor to perform the method of any one of claims 1-7.
16. A non-transitory computer readable storage medium having stored thereon computer instructions for causing the computer to perform the method of any of claims 1-7.
17. A computer program product comprising a computer program which, when executed by a processor, implements a method according to any one of claims 1 to 7.
CN202210507919.4A 2022-05-10 2022-05-10 Method, device, equipment, medium and product for evaluating traffic interaction performance Pending CN114973656A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202210507919.4A CN114973656A (en) 2022-05-10 2022-05-10 Method, device, equipment, medium and product for evaluating traffic interaction performance

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202210507919.4A CN114973656A (en) 2022-05-10 2022-05-10 Method, device, equipment, medium and product for evaluating traffic interaction performance

Publications (1)

Publication Number Publication Date
CN114973656A true CN114973656A (en) 2022-08-30

Family

ID=82982221

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202210507919.4A Pending CN114973656A (en) 2022-05-10 2022-05-10 Method, device, equipment, medium and product for evaluating traffic interaction performance

Country Status (1)

Country Link
CN (1) CN114973656A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115412883A (en) * 2022-08-31 2022-11-29 重庆交通大学 Intelligent network connection over-the-horizon driving auxiliary system based on 5G position sharing

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110285977A (en) * 2019-03-27 2019-09-27 北京智能车联产业创新中心有限公司 Test method, device, equipment and the storage medium of automatic driving vehicle
US20210165409A1 (en) * 2019-12-03 2021-06-03 Mitsubishi Electric Research Laboratories, Inc. Adaptive Control of Autonomous or Semi-Autonomous Vehicle
EP3846035A1 (en) * 2020-01-02 2021-07-07 Beijing Baidu Netcom Science And Technology Co., Ltd. Method, electronic device and storage medium for testing autonomous driving system
CN113687600A (en) * 2021-10-21 2021-11-23 中智行科技有限公司 Simulation test method, simulation test device, electronic equipment and storage medium
WO2022000958A1 (en) * 2020-06-29 2022-01-06 中睿智能交通技术有限公司 V2x-based method for evaluating activation probability of collision warning at intersection without traffic lights
CN114282776A (en) * 2021-12-07 2022-04-05 清华大学 Method, device, equipment and medium for cooperatively evaluating automatic driving safety of vehicle and road

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110285977A (en) * 2019-03-27 2019-09-27 北京智能车联产业创新中心有限公司 Test method, device, equipment and the storage medium of automatic driving vehicle
US20210165409A1 (en) * 2019-12-03 2021-06-03 Mitsubishi Electric Research Laboratories, Inc. Adaptive Control of Autonomous or Semi-Autonomous Vehicle
EP3846035A1 (en) * 2020-01-02 2021-07-07 Beijing Baidu Netcom Science And Technology Co., Ltd. Method, electronic device and storage medium for testing autonomous driving system
WO2022000958A1 (en) * 2020-06-29 2022-01-06 中睿智能交通技术有限公司 V2x-based method for evaluating activation probability of collision warning at intersection without traffic lights
CN113687600A (en) * 2021-10-21 2021-11-23 中智行科技有限公司 Simulation test method, simulation test device, electronic equipment and storage medium
CN114282776A (en) * 2021-12-07 2022-04-05 清华大学 Method, device, equipment and medium for cooperatively evaluating automatic driving safety of vehicle and road

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115412883A (en) * 2022-08-31 2022-11-29 重庆交通大学 Intelligent network connection over-the-horizon driving auxiliary system based on 5G position sharing

Similar Documents

Publication Publication Date Title
US10642268B2 (en) Method and apparatus for generating automatic driving strategy
JP7371157B2 (en) Vehicle monitoring method, device, electronic device, storage medium, computer program, cloud control platform and roadway coordination system
CN111680362B (en) Automatic driving simulation scene acquisition method, device, equipment and storage medium
US20220035733A1 (en) Method and apparatus for checking automatic driving algorithm, related device and storage medium
CN114415628A (en) Automatic driving test method and device, electronic equipment and storage medium
WO2022090800A1 (en) Systems and methods for evaluating domain-specific navigation system capabilities
CN112634611A (en) Method, device, equipment and storage medium for identifying road conditions
CN109910880B (en) Vehicle behavior planning method and device, storage medium and terminal equipment
CN115273477B (en) Intersection driving suggestion pushing method, device, system and electronic equipment
CN114120650A (en) Method and device for generating test result
CN115601983A (en) Method, device, equipment and storage medium for determining cycle duration of traffic signal lamp
CN114360239A (en) Traffic prediction method and system for multilayer space-time traffic knowledge map reconstruction
CN112559371A (en) Automatic driving test method and device and electronic equipment
CN114973656A (en) Method, device, equipment, medium and product for evaluating traffic interaction performance
US11673577B2 (en) System and methods of adaptive relevancy prediction for autonomous driving
CN114720148B (en) Method, device, equipment and storage medium for determining vehicle perception capability
CN114582125B (en) Method, device, equipment and storage medium for identifying road traffic direction
CN114596704B (en) Traffic event processing method, device, equipment and storage medium
CN115973190A (en) Decision-making method and device for automatically driving vehicle and electronic equipment
CN113850297B (en) Road data monitoring method and device, electronic equipment and storage medium
CN115214722A (en) Automatic driving method, device, electronic equipment, storage medium and vehicle
CN111563046B (en) Method and device for generating information
CN114620039A (en) Trajectory correction method and equipment, cloud control platform and automatic driving vehicle
CN114333416A (en) Vehicle risk early warning method and device based on neural network and automatic driving vehicle
CN113610059A (en) Vehicle control method and device based on regional assessment and intelligent traffic management system

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