CN111882924A - Vehicle testing system, driving behavior judgment control method and accident early warning method - Google Patents
Vehicle testing system, driving behavior judgment control method and accident early warning method Download PDFInfo
- Publication number
- CN111882924A CN111882924A CN202010736779.9A CN202010736779A CN111882924A CN 111882924 A CN111882924 A CN 111882924A CN 202010736779 A CN202010736779 A CN 202010736779A CN 111882924 A CN111882924 A CN 111882924A
- Authority
- CN
- China
- Prior art keywords
- vehicle
- driving
- behavior
- test
- aggressive
- 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.)
- Withdrawn
Links
Images
Classifications
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/16—Anti-collision systems
- G08G1/165—Anti-collision systems for passive traffic, e.g. including static obstacles, trees
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01M—TESTING STATIC OR DYNAMIC BALANCE OF MACHINES OR STRUCTURES; TESTING OF STRUCTURES OR APPARATUS, NOT OTHERWISE PROVIDED FOR
- G01M17/00—Testing of vehicles
- G01M17/007—Wheeled or endless-tracked vehicles
-
- 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
- G08G1/0137—Measuring and analyzing of parameters relative to traffic conditions for specific applications
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/01—Detecting movement of traffic to be counted or controlled
- G08G1/052—Detecting movement of traffic to be counted or controlled with provision for determining speed or overspeed
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/09—Arrangements for giving variable traffic instructions
- G08G1/0962—Arrangements for giving variable traffic instructions having an indicator mounted inside the vehicle, e.g. giving voice messages
- G08G1/0967—Systems involving transmission of highway information, e.g. weather, speed limits
- G08G1/096708—Systems involving transmission of highway information, e.g. weather, speed limits where the received information might be used to generate an automatic action on the vehicle control
- G08G1/096725—Systems involving transmission of highway information, e.g. weather, speed limits where the received information might be used to generate an automatic action on the vehicle control where the received information generates an automatic action on the vehicle control
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/16—Anti-collision systems
- G08G1/164—Centralised systems, e.g. external to vehicles
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/16—Anti-collision systems
- G08G1/166—Anti-collision systems for active traffic, e.g. moving vehicles, pedestrians, bikes
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/09—Arrangements for giving variable traffic instructions
- G08G1/0962—Arrangements for giving variable traffic instructions having an indicator mounted inside the vehicle, e.g. giving voice messages
- G08G1/0967—Systems involving transmission of highway information, e.g. weather, speed limits
- G08G1/096766—Systems involving transmission of highway information, e.g. weather, speed limits where the system is characterised by the origin of the information transmission
- G08G1/096775—Systems involving transmission of highway information, e.g. weather, speed limits where the system is characterised by the origin of the information transmission where the origin of the information is a central station
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/09—Arrangements for giving variable traffic instructions
- G08G1/0962—Arrangements for giving variable traffic instructions having an indicator mounted inside the vehicle, e.g. giving voice messages
- G08G1/0967—Systems involving transmission of highway information, e.g. weather, speed limits
- G08G1/096766—Systems involving transmission of highway information, e.g. weather, speed limits where the system is characterised by the origin of the information transmission
- G08G1/096783—Systems involving transmission of highway information, e.g. weather, speed limits where the system is characterised by the origin of the information transmission where the origin of the information is a roadside individual element
Abstract
The invention relates to a vehicle test system, a driving behavior judgment control method and an accident early warning method based on a racing yard, wherein the vehicle test system comprises center end equipment and road end equipment, the road end equipment is arranged in the racing yard and comprises an auxiliary positioning facility, a communication facility, a traffic sign marking, a traffic control and induction facility, a road end calculation facility and a traffic perception facility, and the test system realizes an open environment road test of an automatic driving vehicle and a mixed traffic flow test of a closed area of the racing track based on the racing yard; the judgment control method is used for constructing and training an aggressive driving scene understanding model based on data obtained by a vehicle aggressive driving test in a high-speed driving environment of a vehicle test system; and acquiring driving behavior data and driving scene information of the automatic driving vehicle in real time, and realizing judgment and control of the aggressive driving behavior of the automatic driving vehicle based on the aggressive driving scene understanding model. Compared with the prior art, the invention has the advantages of various testing functions, simple realization and the like.
Description
Technical Field
The invention relates to the technical field of automobile testing, in particular to a vehicle testing system based on a racing car park, a driving behavior judgment control method and an accident early warning method.
Background
Along with the development of intelligent automobile technology, human trip modes face huge changes, road traffic environments are increasingly complex, the complexity of the road traffic environments with both driving and unmanned driving is increasingly improved, and the technology for finely managing and controlling road traffic safety and improving the skills of drivers in complex scenes become urgent requirements for long-term existence and under complex scenes.
In the automotive industry, all technologies are going to be produced in volume from laboratories and all need to be validated. In addition to efforts of various technical solution companies, including but not limited to OEMs (original equipment manufacturers), automatic driving companies, conventional automatic driving vehicles are going to automatic driving, and the experimental results are required to be continuously tested and symmetrically debugged and optimized. Before an autonomous vehicle can be taken on the road formally, a targeted test is required to prove the operational safety of the autonomous vehicle. Drive testing is undoubtedly the most direct way, but because of the weight and speed of the autonomous vehicle, there are significant safety hazards to testing in practical scenarios, especially before the technology is not yet mature, the safety hazards are greater. However, if there is no actual drive test, the updating and upgrading of the technology is very difficult.
But at present, the test road suitable for the high-speed limit scene test is not available.
In addition, the accuracy of prediction of the driving behavior of the current autonomous vehicle is not high enough, and further improvement is required particularly in a high-speed driving environment.
Disclosure of Invention
One of the objectives of the present invention is to provide a vehicle testing system based on a racing car park, which has various testing functions and is simple to implement, in order to overcome the defects of the prior art.
The second purpose of the present invention is to provide a highly reliable determination control method for aggressive driving behavior of an autonomous vehicle to overcome the above-mentioned drawbacks of the prior art, which is helpful for improving the determination and processing capabilities of an autonomous device during meeting or following an aggressive driving in a high-speed driving environment.
The invention also aims to overcome the defects of the prior art and provide an automatic vehicle driving accident early warning method with high reliability
One of the purposes of the invention can be realized by the following technical scheme:
a vehicle testing system based on a racing yard, comprising center-end equipment and end-of-road equipment, the end-of-road equipment being disposed within the racing yard, comprising:
the auxiliary positioning facility is used for positioning and updating the automatic driving vehicle;
the communication facility is respectively connected with the automatic driving vehicle, the central terminal equipment and the auxiliary positioning device, realizes information exchange with the automatic driving vehicle and the central terminal equipment, and realizes clock synchronization based on the auxiliary positioning device;
the traffic sign marking is arranged on the racing track;
a traffic control and guidance facility connected to the autonomous vehicle and transmitting traffic control and guidance information to the autonomous vehicle;
the road end computing facility is used for carrying out data processing and computing based on the real-time data collected on site;
the traffic sensing facility is used for sensing the traffic environment state information and transmitting the traffic environment state information to the road end computing facility or the central end equipment;
the test system realizes the open environment road test of the automatic driving vehicle and the mixed traffic flow test of the track closed area based on the track field.
Further, the open environment road test comprises a no-signal intersection vehicle road cooperative test, a signal intersection vehicle speed guide test, a tunnel test, a long downhill test and a roundabout test;
the race track closed area mixed traffic flow test is realized based on a mixed traffic flow environment in a high-speed running state, and the mixed traffic flow environment comprises vehicle following, lane changing, formation, lane keeping and tracking, cooperative self-adaptive cruise, cooperative emergency braking, cooperative lane changing, remote control driving and manual simulation intervention environment.
Further, the center-end equipment comprises an automatic driving monitoring and service center and a high-precision map generation and storage facility.
Further, the road end equipment also comprises energy supply and lighting facilities.
The second purpose of the invention can be realized by the following technical scheme:
a method for judging and controlling aggressive driving behavior of an automatically driven vehicle comprises the following steps:
s11) carrying out vehicle aggressive driving test under high-speed driving environment by adopting the vehicle testing system based on the racing yard as claimed in claim 1, and acquiring behavior characteristics and corresponding driving scene information of the vehicle corresponding to each aggressive driving behavior;
s12) taking the data obtained by the test in the step S11) as training data, and constructing and training an aggressive driving scene understanding model;
s13) acquiring the driving behavior data and the driving scene information of the automatic driving vehicle in real time, and realizing judgment and control of the aggressive driving behavior of the automatic driving vehicle based on the aggressive driving scene understanding model;
the aggressive driving scene understanding model comprises a behavior network and an understanding network, wherein the behavior network is used for judging the type of the next aggressive driving behavior and the attention mechanism target, the understanding network is used for obtaining semantic description of the attention mechanism target, and the automatic driving vehicle controls the driving behavior of the automatic driving vehicle based on the type of the next aggressive driving behavior and the semantic description of the attention mechanism target.
Further, the vehicles for carrying out the vehicle aggressive driving test comprise automatic driving vehicles and manned vehicles.
Further, the driving scenario information includes at least one of: actual travel trajectory, static obstacle information, dynamic obstacle information, and road information.
Further, the behavior network includes:
the search classification sub-network is used for obtaining behavior characteristic information according to the driving behavior data obtained in real time, and performing search classification based on the behavior characteristic information to obtain an aggressive driving behavior category;
and the attention sub-network is used for determining the target objects according to the driving behavior types and the corresponding driving scene information, identifying the distance of each target object by using a responsibility sensitive model, and marking the target objects with the distance smaller than a preset value with attention labels to generate the attention mechanism target.
Further, the understanding network includes:
the convolutional neural subnetwork is used for performing parallel convolution processing on different frames and extracting the target characteristics of the attention mechanism target;
and the long-term and short-term memory sub-network is used for allocating different weights to each frame based on the target characteristics and the image information, capturing the action characteristics of the attention mechanism target by means of an optical flow method, and acquiring semantic description of the attention mechanism target.
The third purpose of the invention can be realized by the following technical scheme:
an accident early warning method for an automatic driving vehicle, which comprises the following steps:
s21) carrying out vehicle aggressive driving test under high-speed driving environment by adopting the vehicle testing system based on the racing yard as claimed in claim 1, obtaining behavior characteristics and corresponding driving scene information corresponding to each aggressive driving behavior of the vehicle, and forming an aggressive driving behavior database;
s22) acquiring real-time behavior data of the vehicle based on the real-time acquired vehicle driving data, road information and weather conditions, comparing the real-time behavior data of the vehicle with existing data in an aggressive driving behavior database, judging whether collision risks exist, if not, returning to the step S22) for next round of judgment, and if so, generating an early warning signal and sending the early warning signal to a traffic control and guidance facility.
Compared with the prior art, the invention has the following beneficial effects:
1. the invention organically combines the car racing field with the intelligent network technology, sets a comprehensive test scene aiming at the characteristics of the car racing field, effectively utilizes the existing car racing field, reduces the test cost and can realize the vehicle test in scenes such as high-speed limit and the like.
2. The method can effectively obtain vehicle aggressive driving test data in a high-speed driving environment based on the constructed test system, thereby constructing aggressive driving scene understanding of the automatic driving equipment, effectively improving the reliable control of aggressive driving behaviors of the automatic driving equipment and being beneficial to the construction of an aggressive driving behavior database in the high-speed environment.
3. The invention can quickly and accurately realize the early warning of the car race accident according to the aggressive driving behavior database in the high-speed environment.
Drawings
FIG. 1 is a schematic view of a test of an open environment scene at the periphery of a track in a track field under the test system of the present invention;
FIG. 2 is a schematic view of a track closed area scene test in a racing track area under the test system of the present invention;
FIG. 3 is a schematic view of an understanding process of an aggressive driving scenario of the present invention;
fig. 4 is a schematic diagram of the accident warning process of the present invention.
Detailed Description
The invention is described in detail below with reference to the figures and specific embodiments. The present embodiment is implemented on the premise of the technical solution of the present invention, and a detailed implementation manner and a specific operation process are given, but the scope of the present invention is not limited to the following embodiments.
Example 1
The embodiment provides a vehicle testing system based on a racing yard, which is realized by deploying intelligent network connection auxiliary facilities in the whole racing yard, and comprises center end equipment and road end equipment, wherein the road end equipment is arranged in the racing yard and comprises an auxiliary positioning facility, a communication facility, a traffic sign marking line, a traffic control and guidance facility, a road end calculation facility and a traffic perception facility, and the specific setting type, the number and the position of the road end equipment can be determined according to specific conditions; the center-end equipment comprises an automatic driving monitoring and service center and a high-precision map generation and storage facility.
The functions of the above facilities mainly include:
1) and the automatic driving monitoring and service center is used for gathering, processing and managing the automatic driving of the road in the jurisdiction and the service related information of the automatic driving.
2) And the high-precision map generation storage facility is used for storing static traffic data and dynamic traffic data of the roads under jurisdiction.
3) Roadside assistance-localization real-time facilities, to relying on the automatic vehicle of inertial navigation can continuously accumulate the positioning error in the course of going, when positioning error surpassed 10m, normal operating just can't be guaranteed to all kinds of basic positioning navigation functions, need assistance-localization real-time system to carry out the calibration that the precision is higher to it, in order to eliminate the accumulative error.
4) The road-end communication facility not only has the functions of receiving and sending wireless signals, completes the information exchange between the automatic driving vehicle and the road-end facility and between the road-end facility and the intelligent network monitoring and service center, but also can receive clock signals provided by the high-precision positioning facility and is used for the clock synchronization of the road-end communication facility.
5) The traffic sign line is used for indicating traffic prohibition, restriction and compliance conditions of a road for an automatic driving vehicle and informing road conditions and traffic condition information.
6) The traffic control and guidance facility can be networked with the automatic driving vehicle, can issue traffic control and guidance information to the vehicle through a wireless communication network, can be networked with the road end computing facility, receives the traffic control and guidance information sent by the road end computing facility, and sends the information to the automatic driving vehicle.
7) The traffic perception facility can collect information such as traffic running conditions, traffic events, road meteorological environment, infrastructure states and the like, and can send the perceived information to a road end computing facility or an automatic driving monitoring and service center in a wired or wireless mode.
8) And the road end computing facility consists of a data processing and controlling unit, a data storage unit and a communication interface, and is used for completing the collection of the relevant information of automatic driving and the on-site rapid processing.
In another embodiment, the end-of-road equipment further includes energy supply and lighting facilities to provide a desired energy supply and a desired lighting environment for the autonomous vehicle and associated ancillary facilities.
In another embodiment, network security facilities including software and hardware facilities are deployed in both the center-end equipment and the road-end equipment, so that the hardware, software and data of related systems are protected from being damaged, changed and leaked in the information exchange process between the automatic driving vehicle and the accessory facilities and between the accessory facilities.
The test system can realize the open environment road test of the automatic driving vehicle and the mixed traffic flow test of the track closed area based on the track field.
As shown in fig. 1, since the peripheral loop of the race track is restricted by the terrain of the track, the longitudinal slope of the road is large, the sight distance is limited, and the track is provided with a plurality of tunnels arranged for passing through the track to enter and exit the area of the race track, diversified open environment road tests can be provided for the automatic driving vehicles, and a vehicle-road cooperative test at a non-signal intersection, a vehicle speed guide test at a signal intersection, a tunnel test, a long downhill test, a roundabout test and the like can be provided.
As shown in fig. 2, in addition, in the track closure area, a mixed traffic flow test scene in a high-speed running state of the vehicle can be constructed by the automatic driving vehicle and the professional racing driver, and test environments such as vehicle following, lane changing, formation, lane keeping and tracking, cooperative adaptive cruise, cooperative emergency braking, cooperative lane changing test, remote control driving, manual simulation intervention and the like are provided.
Example 2
In this embodiment, the test system according to embodiment 1 is used to implement a method for determining and controlling aggressive driving behavior of an autonomous vehicle, and the method includes the following steps:
s11) performing a vehicle aggressive driving test in a high-speed driving environment by using the vehicle testing system based on the racing yard as described in embodiment 1, and acquiring behavior characteristics and corresponding driving scenario information of the vehicle corresponding to each aggressive driving behavior;
s12) taking the data obtained by the test in the step S11) as training data, and constructing and training an aggressive driving scene understanding model;
s13) acquiring driving behavior data and driving scene information of the automatic driving vehicle in real time, and realizing judgment and control of the automatic driving vehicle aggressive driving behavior based on the aggressive driving scene understanding model, wherein the driving scene information comprises at least one of the following information: actual travel trajectory, static obstacle information, dynamic obstacle information, and road information.
According to the method, in view of the characteristics of the racing car park, a high-speed mixed traffic flow test scene of shared roads of unmanned driving and manned driving is constructed in the racing car park. The intelligent network connection end facility arranged on the racing car park can better and more quickly assist the automatic driving equipment to finish understanding scenes. For example, the road end sensing equipment can directly transmit the traffic running conditions, traffic events, road meteorological environment and infrastructure states around the vehicle to the automatic driving equipment more accurately and quickly; the automatic driving equipment needing learning scenes does not need to be arranged on the test vehicles, and can be arranged in the road end computing facility, so that the cost of arranging the automatic driving equipment on each test vehicle is reduced, and the compatibility of the system is improved. After data of a high-speed mixed traffic flow test are obtained, the understanding level of the automatic driving equipment on the aggressive driving scene is trained through specific aggressive driving scene understanding, so that the construction of an aggressive driving behavior database in a high-speed environment is completed, the category of the aggressive driving behavior and corresponding driving scene information are identified on the basis of real-time data acquisition, the driving scene understanding is completed on the basis of a marked target object, and the autonomous aggressive driving behavior control of the automatic driving vehicle is realized.
The aggressive driving behaviors comprise driving behaviors such as rapid acceleration, rapid deceleration, rapid turning, rapid braking, rapid overtaking and the like. Each type of aggressive driving behaviors has different behavior characteristics, the values of the advancing direction and the transverse linear acceleration of the motor vehicle are obtained according to an acceleration sensor arranged on the vehicle to obtain corresponding behavior characteristics, so that the driving behaviors are classified, different analyses can be carried out according to different classes, different target objects needing attention under different aggressive driving behaviors are determined, understanding of a driving scene is completed according to the state of the real-time target objects, and the driving behaviors of the automatic driving vehicle are controlled.
According to the aggressive driving behaviors and the categories to which the aggressive driving behaviors belong, the corresponding target objects can be identified from the driving scenes corresponding to the aggressive driving behaviors. For example, taking a sharp turn as an example, it is possible to select a target object such as a front intersection, a traffic light, and a front and rear vehicle, which need to be referred to for a sharp turn corresponding to such a sharp driving behavior. The target objects corresponding to various types of aggressive driving behaviors are peripheral driving scenes for the automatic driving equipment, and the state change of the target objects in a period of time can affect the driving behaviors. The driving scene of the automatic driving equipment can be shown by combining the target objects.
As shown in fig. 3, the understanding of the driving scenario can be realized by means of a behavior network and an understanding network, wherein the behavior network can comprise a search classification sub-network and an attention sub-network, and the understanding network can be realized by a long-short term memory sub-network. The behavior network input end can input driving behavior data, specifically including acceleration of the vehicle, steering wheel angle and the like. Because sharp driving behaviors such as rapid acceleration, rapid deceleration, sharp turn, rapid braking, rapid overtaking and the like have obvious characteristics, the search and classification can be carried out based on the characteristics, the variation characteristic of the acceleration and the variation characteristic of the steering angle of the steering wheel can be selected as characteristics, and category labels are marked on the sharp driving behavior data, and the category labels comprise at least one of the following: rapid acceleration, rapid deceleration, rapid turning, rapid braking and rapid overtaking. For example, the sharp driving behavior characteristic that the acceleration is suddenly increased and the steering angle of the steering wheel is unchanged is determined as the rapid acceleration; for example, the characteristics of aggressive driving behavior with suddenly small acceleration and unchanged steering angle of a steering wheel are determined as rapid deceleration; for example, the aggressive driving behavior characteristic that the acceleration is suddenly increased and the steering angle of the steering wheel is increased first and then decreased is determined as the rapid overtaking.
According to the type of the aggressive driving behaviors, the attention sub-network is utilized to carry out corresponding attention processing on the driving scene information of various aggressive driving behaviors, target objects are determined based on the aggressive driving behaviors after the attention processing and the driving scene information corresponding to the aggressive driving behaviors, a responsibility sensitive model is utilized to carry out distance recognition on each target object, and the target objects with the distances smaller than a preset value are marked with attention labels. An attention subnetwork utilizes a network node that is set up by selectively focusing on a portion of all information while ignoring other visible information, with an attention mechanism. A responsibility sensitive model is a model that mathematically defines a range of influence. For example, for an aggressive driving behavior of sudden stop, an attention network is utilized to perform attention search to find out whether a front vehicle or a traffic signal lamp exists in the driving direction of the vehicle, and if so, the front vehicle or the traffic signal lamp is directly used as a target object to be marked with an attention label; if the object does not exist, attention is directly added around the vehicle, obstacles around the vehicle are judged according to the responsibility sensitive module, and objects in the influence range are marked. When the vehicle is overtaking suddenly, attention is paid to the front and the side of the vehicle, and objects in an influence range can be marked. Thus, corresponding attention processing can be carried out according to different aggressive driving behaviors.
In addition, in order to more accurately and efficiently analyze the image video data of the target object, the convolution neural sub-network can be used in the understanding network to perform convolution processing on the image frame containing the target object. The specific operation is as follows, the data output after the search classification sub-network and the attention sub-network in the behavior network are processed can be used as the input for understanding the network. The understanding network takes the output of the behavior network as input, the convolutional neural sub-network performs parallel convolution processing on different frames, and the features of the attention mechanism target are extracted to be used as the input of the long-term and short-term memory network. The long-short term memory sub-network assigns different weights to each frame based on the characteristics and the information such as the position in the image, and captures the action characteristics of the attention mechanism target by means of an optical flow method. The final output of the whole understanding network is a semantic description of different attention mechanism targets. In this way, an understanding of the driving scenario is achieved. The convolutional neural sub-network is a feed-forward neural network which comprises convolutional calculation and has a deep structure, can learn pixels and audio and has a stable effect, and can extract images corresponding to a target object and analyze and process characteristics. The long-short term memory subnetwork is a time-cycle neural network suitable for processing and predicting important events of very long intervals and delays in a time series as complex nonlinear units. Optical flow can be used to describe the motion of an observation target caused by motion relative to an observer.
Therefore, the driving scene understanding method shown in fig. 3 effectively learns the aggressive driving behaviors of the human driver, specifically identifies the aggressive driving behaviors and marks corresponding target objects, improves the understanding level of the driving scene of the automatic driving device, and is beneficial to improving the judging and processing capabilities of the automatic driving device in meeting or following the aggressive driving in a high-speed driving environment.
Example 3
As shown in fig. 4, the present embodiment provides an accident pre-warning method for an autonomous vehicle, including the following steps:
s21) performing a vehicle aggressive driving test in a high-speed driving environment by using the vehicle testing system based on the racing yard as described in embodiment 1, obtaining behavior characteristics and corresponding driving scenario information corresponding to each aggressive driving behavior of the vehicle, and forming an aggressive driving behavior database;
s22) acquiring real-time behavior data of the vehicle based on the real-time acquired vehicle driving data, road information and weather conditions, comparing the real-time behavior data of the vehicle with existing data in an aggressive driving behavior database, judging whether collision risks exist, if not, returning to the step S22) to perform next round of judgment, and if so, generating an early warning signal and sending the early warning signal to a traffic control and guidance facility to remind the vehicle of avoiding. If a collision still occurs, an emergency treatment protocol is automatically initiated.
The accident early warning method can be deployed in an automatic driving monitoring and service center.
The foregoing detailed description of the preferred embodiments of the invention has been presented. It should be understood that numerous modifications and variations could be devised by those skilled in the art in light of the present teachings without departing from the inventive concepts. Therefore, the technical solutions available to those skilled in the art through logic analysis, reasoning and limited experiments based on the prior art according to the concept of the present invention should be within the scope of protection defined by the claims.
Claims (10)
1. A vehicle test system based on a racing car park is characterized by comprising center end equipment and road end equipment, wherein the road end equipment is arranged in the racing car park and comprises:
the auxiliary positioning facility is used for positioning and updating the automatic driving vehicle;
the communication facility is respectively connected with the automatic driving vehicle, the central terminal equipment and the auxiliary positioning device, realizes information exchange with the automatic driving vehicle and the central terminal equipment, and realizes clock synchronization based on the auxiliary positioning device;
the traffic sign marking is arranged on the racing track;
a traffic control and guidance facility connected to the autonomous vehicle and transmitting traffic control and guidance information to the autonomous vehicle;
the road end computing facility is used for carrying out data processing and computing based on the real-time data collected on site;
the traffic sensing facility is used for sensing the traffic environment state information and transmitting the traffic environment state information to the road end computing facility or the central end equipment;
the test system realizes the open environment road test of the automatic driving vehicle and the mixed traffic flow test of the track closed area based on the track field.
2. The racing yard-based vehicle testing system of claim 1, wherein the open environment road test comprises a no-signalized intersection vehicle road cooperation test, a signalized intersection vehicle speed guidance test, a tunnel test, a long downhill test, and a roundabout test;
the race track closed area mixed traffic flow test is realized based on a mixed traffic flow environment in a high-speed running state, and the mixed traffic flow environment comprises vehicle following, lane changing, formation, lane keeping and tracking, cooperative self-adaptive cruise, cooperative emergency braking, cooperative lane changing, remote control driving and manual simulation intervention environment.
3. The race track-based vehicle testing system of claim 1, characterized in that the central end equipment comprises an automated driving monitoring and service center and high precision map generation storage facilities.
4. The race track-based vehicle testing system of claim 1, wherein the end-of-road equipment further comprises power and lighting facilities.
5. A method for judging and controlling aggressive driving behavior of an automatically driven vehicle is characterized by comprising the following steps:
s11) carrying out vehicle aggressive driving test under high-speed driving environment by adopting the vehicle testing system based on the racing yard as claimed in claim 1, and acquiring behavior characteristics and corresponding driving scene information of the vehicle corresponding to each aggressive driving behavior;
s12) taking the data obtained by the test in the step S11) as training data, and constructing and training an aggressive driving scene understanding model;
s13) acquiring the driving behavior data and the driving scene information of the automatic driving vehicle in real time, and realizing judgment and control of the aggressive driving behavior of the automatic driving vehicle based on the aggressive driving scene understanding model;
the aggressive driving scene understanding model comprises a behavior network and an understanding network, wherein the behavior network is used for judging the type of the next aggressive driving behavior and the attention mechanism target, the understanding network is used for obtaining semantic description of the attention mechanism target, and the automatic driving vehicle controls the driving behavior of the automatic driving vehicle based on the type of the next aggressive driving behavior and the semantic description of the attention mechanism target.
6. The aggressive driving behavior judgment control method for an autonomous vehicle according to claim 5, characterized in that vehicles subjected to the aggressive driving test for vehicles include an autonomous vehicle and a manned vehicle.
7. The aggressive driving behavior determination control method of an autonomous vehicle according to claim 5, characterized in that the driving scenario information includes at least one of: actual travel trajectory, static obstacle information, dynamic obstacle information, and road information.
8. The aggressive driving behavior determination control method for an autonomous vehicle according to claim 7, characterized in that the behavior network includes:
the search classification sub-network is used for obtaining behavior characteristic information according to the driving behavior data obtained in real time, and performing search classification based on the behavior characteristic information to obtain an aggressive driving behavior category;
and the attention sub-network is used for determining the target objects according to the driving behavior types and the corresponding driving scene information, identifying the distance of each target object by using a responsibility sensitive model, and marking the target objects with the distance smaller than a preset value with attention labels to generate the attention mechanism target.
9. The determination control method for aggressive driving behavior of an autonomous vehicle as claimed in claim 7, wherein said understanding network comprises:
the convolutional neural subnetwork is used for performing parallel convolution processing on different frames and extracting the target characteristics of the attention mechanism target;
and the long-term and short-term memory sub-network is used for allocating different weights to each frame based on the target characteristics and the image information, capturing the action characteristics of the attention mechanism target by means of an optical flow method, and acquiring semantic description of the attention mechanism target.
10. An accident early warning method for an autonomous vehicle, the method comprising the steps of:
s21) carrying out vehicle aggressive driving test under high-speed driving environment by adopting the vehicle testing system based on the racing yard as claimed in claim 1, obtaining behavior characteristics and corresponding driving scene information corresponding to each aggressive driving behavior of the vehicle, and forming an aggressive driving behavior database;
s22) acquiring real-time behavior data of the vehicle based on the real-time acquired vehicle driving data, road information and weather conditions, comparing the real-time behavior data of the vehicle with existing data in an aggressive driving behavior database, judging whether collision risks exist, if not, returning to the step S22) for next round of judgment, and if so, generating an early warning signal and sending the early warning signal to a traffic control and guidance facility.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010736779.9A CN111882924A (en) | 2020-07-28 | 2020-07-28 | Vehicle testing system, driving behavior judgment control method and accident early warning method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010736779.9A CN111882924A (en) | 2020-07-28 | 2020-07-28 | Vehicle testing system, driving behavior judgment control method and accident early warning method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111882924A true CN111882924A (en) | 2020-11-03 |
Family
ID=73201134
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010736779.9A Withdrawn CN111882924A (en) | 2020-07-28 | 2020-07-28 | Vehicle testing system, driving behavior judgment control method and accident early warning method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111882924A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112837540A (en) * | 2020-12-31 | 2021-05-25 | 南斗六星系统集成有限公司 | Intelligent site management system and method |
| CN114333419A (en) * | 2021-12-28 | 2022-04-12 | 杭州海康威视系统技术有限公司 | Traffic conflict determination method, device, equipment and readable storage medium |
| CN115798080A (en) * | 2023-02-02 | 2023-03-14 | 江苏天一航空工业股份有限公司 | Method for testing cooperative exit mechanism of civil aviation airport vehicle and road |
Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR100440110B1 (en) * | 1997-11-28 | 2004-09-24 | 현대자동차주식회사 | Test system and control method thereof for unmanned vehicle for automatically operating test car in mechanical method without driver |
| CN103335853A (en) * | 2013-07-18 | 2013-10-02 | 中国科学院自动化研究所 | Unmanned driving vehicle cognitive competence testing system and method |
| CN105699095A (en) * | 2016-01-27 | 2016-06-22 | 常州加美科技有限公司 | A test method for a remotely piloted vehicle (PRV) |
| US20160334789A1 (en) * | 2015-05-11 | 2016-11-17 | Hyundai Motor Company | Driving Test System for a Moving Object |
| CN106153352A (en) * | 2016-07-04 | 2016-11-23 | 江苏大学 | A kind of automatic driving vehicle test and verification platform and method of testing thereof |
| CN106706342A (en) * | 2017-02-17 | 2017-05-24 | 特路（北京）科技有限公司 | Testing method for continuous crossroad passing ability of automatic vehicle and testing field |
| CN107843440A (en) * | 2017-10-30 | 2018-03-27 | 青岛慧拓智能机器有限公司 | A kind of automatic driving vehicle Performance Test System and method |
| CN108645628A (en) * | 2018-05-14 | 2018-10-12 | 公安部交通管理科学研究所 | Autonomous driving vehicle based on driving path technical ability tests system |
| CN109324539A (en) * | 2018-08-28 | 2019-02-12 | 山东省科学院自动化研究所 | The intelligent control platform and method of a kind of automatic Pilot closed test field |
| CN110597711A (en) * | 2019-08-26 | 2019-12-20 | 湖南大学 | Automatic driving test case generation method based on scene and task |
| CN110606093A (en) * | 2019-09-26 | 2019-12-24 | 中国第一汽车股份有限公司 | Vehicle performance evaluation method, device, equipment and storage medium |
| CN111179585A (en) * | 2018-11-09 | 2020-05-19 | 上海汽车集团股份有限公司 | Site testing method and device for automatic driving vehicle |
-
2020
- 2020-07-28 CN CN202010736779.9A patent/CN111882924A/en not_active Withdrawn
Patent Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR100440110B1 (en) * | 1997-11-28 | 2004-09-24 | 현대자동차주식회사 | Test system and control method thereof for unmanned vehicle for automatically operating test car in mechanical method without driver |
| CN103335853A (en) * | 2013-07-18 | 2013-10-02 | 中国科学院自动化研究所 | Unmanned driving vehicle cognitive competence testing system and method |
| US20160334789A1 (en) * | 2015-05-11 | 2016-11-17 | Hyundai Motor Company | Driving Test System for a Moving Object |
| CN105699095A (en) * | 2016-01-27 | 2016-06-22 | 常州加美科技有限公司 | A test method for a remotely piloted vehicle (PRV) |
| CN106153352A (en) * | 2016-07-04 | 2016-11-23 | 江苏大学 | A kind of automatic driving vehicle test and verification platform and method of testing thereof |
| CN106706342A (en) * | 2017-02-17 | 2017-05-24 | 特路（北京）科技有限公司 | Testing method for continuous crossroad passing ability of automatic vehicle and testing field |
| CN107843440A (en) * | 2017-10-30 | 2018-03-27 | 青岛慧拓智能机器有限公司 | A kind of automatic driving vehicle Performance Test System and method |
| CN108645628A (en) * | 2018-05-14 | 2018-10-12 | 公安部交通管理科学研究所 | Autonomous driving vehicle based on driving path technical ability tests system |
| CN109324539A (en) * | 2018-08-28 | 2019-02-12 | 山东省科学院自动化研究所 | The intelligent control platform and method of a kind of automatic Pilot closed test field |
| CN111179585A (en) * | 2018-11-09 | 2020-05-19 | 上海汽车集团股份有限公司 | Site testing method and device for automatic driving vehicle |
| CN110597711A (en) * | 2019-08-26 | 2019-12-20 | 湖南大学 | Automatic driving test case generation method based on scene and task |
| CN110606093A (en) * | 2019-09-26 | 2019-12-24 | 中国第一汽车股份有限公司 | Vehicle performance evaluation method, device, equipment and storage medium |
Non-Patent Citations (5)
| Title |
|---|
| 刘天洋 等: "智能网联汽车试验场发展现状与建设建议", 《汽车技术》 * |
| 左任婧 等: "国内外智能网联汽车试验场的发展现状", 《北京汽车》 * |
| 李骁驰 等: "面向智能网联交通系统的模块化柔性试验场", 《中国公路学报》 * |
| 王科俊 等: "深度学习在无人驾驶汽车领域应用的研究进展", 《智能系统学报》 * |
| 赵鹏超 等: "智能网联汽车封闭试验场综合管理研究", 《科学技术创新》 * |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112837540A (en) * | 2020-12-31 | 2021-05-25 | 南斗六星系统集成有限公司 | Intelligent site management system and method |
| CN114333419A (en) * | 2021-12-28 | 2022-04-12 | 杭州海康威视系统技术有限公司 | Traffic conflict determination method, device, equipment and readable storage medium |
| CN115798080A (en) * | 2023-02-02 | 2023-03-14 | 江苏天一航空工业股份有限公司 | Method for testing cooperative exit mechanism of civil aviation airport vehicle and road |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| EP3533681B1 (en) | Method for detecting safety of driving behavior, apparatus and storage medium | |
| CN111179585B (en) | Site testing method and device for automatic driving vehicle | |
| CN111540237B (en) | Method for automatically generating vehicle safety driving guarantee scheme based on multi-data fusion | |
| CN107782564B (en) | Automatic driving vehicle evaluation system and method | |
| CN111781855B (en) | Traffic is automatic driving simulation system in ring | |
| CN107843440B (en) | System and method for testing performance of automatic driving vehicle | |
| US10957201B2 (en) | System and method for relative positioning based safe autonomous driving | |
| CN111524362B (en) | Vehicle safety driving guarantee system and method based on multi-data fusion | |
| CN111473980B (en) | Intelligent automobile automatic driving capability test system | |
| CN112700470B (en) | Target detection and track extraction method based on traffic video stream | |
| US11836985B2 (en) | Identifying suspicious entities using autonomous vehicles | |
| CN111524357A (en) | Method for fusing multiple data required for safe driving of vehicle | |
| CN111882924A (en) | Vehicle testing system, driving behavior judgment control method and accident early warning method | |
| CN114077541A (en) | Method and system for validating automatic control software for an autonomous vehicle | |
| CN110356412A (en) | The method and apparatus that automatically rule for autonomous driving learns | |
| CN113511204B (en) | Vehicle lane changing behavior identification method and related equipment | |
| CN109491380A (en) | A kind of intelligent automobile Unmanned Systems and application method | |
| CN111845787A (en) | Lane change intention prediction method based on LSTM | |
| CN113851017A (en) | Pedestrian and vehicle identification and early warning multifunctional system based on road side RSU | |
| CN112835790A (en) | Test method, device, equipment and medium for automatic driving software | |
| CN115165387A (en) | Control method, device and system for testing performance of automatic driving whole vehicle | |
| CN116686028A (en) | Driving assistance method and related equipment | |
| US10953871B2 (en) | Transportation infrastructure communication and control | |
| CN110765224A (en) | Processing method of electronic map, vehicle vision repositioning method and vehicle-mounted equipment | |
| CN116135654A (en) | Vehicle running speed generation method and related equipment |
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 | ||
| WW01 | Invention patent application withdrawn after publication |
Application publication date: 20201103 |
|
| WW01 | Invention patent application withdrawn after publication |