KR102579590B1 - Autonomous Vehicle ability evaluation scenario generation system based on the Road Traffic Act - Google Patents

Abstract

A road traffic law-based autonomous driving ability evaluation scenario generation system is launched. The disclosed road traffic law-based self-driving driving ability evaluation scenario generation system is a road traffic law-based self-driving driving ability evaluation scenario generation system applied to a virtual environment-based self-driving car evaluation system, and is based on road traffic law-based self-driving driving ability evaluation scenario generation system that provides The autonomous vehicle's operating area, operating section, operating road information at the operating point, operating road signals, speed information, and precise map information are received and identified, analyzed, and extracted based on the received information to determine the operating design area ( An information receiving unit that creates an ODD (Operational Design Domain); A single scenario generator that collects driving road information and performance data of the autonomous vehicle from the information receiver, reflects traffic law evaluation items, and converts them into ASAM OpenSCENARIO format to generate a single scenario; Based on the operation design area created by the information receiver, physical elements including road environment and facilities in the virtual environment are verified, and logical elements of vehicles operating on the road other than the evaluation vehicle are verified to determine whether the scenario can be performed. A scenario verification unit that determines and generates a verified verification scenario; A scenario operation DB management unit that collects and analyzes traffic law data applicable to the driving section from the information receiver and performs data management to create a complex scenario by layering the verification scenario data into two or more layers; It is characterized in that it includes a complex scenario generation unit that generates a complex scenario through the scenario operation DB management unit and transmits it to a platform for driving evaluation of an autonomous vehicle.

Description

{Autonomous Vehicle ability evaluation scenario generation system based on the Road Traffic Act}

The present invention relates to a system for generating scenarios for evaluating the driving ability of autonomous vehicles. More specifically, the relevant laws are analyzed to evaluate the safe driving ability of self-driving cars based on the road traffic laws that all road users must comply with, and the self-driving cars comply with them. This is about a road traffic law-based autonomous driving ability evaluation scenario generation system that derives the items and evaluation elements to be done and creates an evaluation scenario using an overseas standard scenario format.

Recently, as the automobile industry and various sensing technologies have advanced and communication infrastructure has expanded, interest in and research on self-driving vehicles that autonomously drive to their destination without driver intervention have been actively conducted.

Autonomous vehicles (autonomous vehicles) use various cameras, sensors, and high-precision maps to recognize, judge, and act on surrounding vehicles, pedestrians, and objects, and control the vehicle without driver intervention to autonomously drive to a set destination. It means a vehicle that can.

If the safety of these self-driving vehicles is not secured, it can lead to a major casualty accident, so it is essential to evaluate and test how quickly and accurately they can recognize and respond to various unexpected situations and events.

The evaluation of autonomous vehicles consists of virtual verification, which evaluates in a virtual driving environment that virtualizes actual driving situations, and actual vehicle verification, which evaluates while driving on actual roads.

In order to perform virtual verification, a virtual road environment based on a real road is required, and in this virtual road environment, traffic signals such as traffic lights, surrounding vehicles, and events such as lane changes, stops, acceleration, and deceleration of surrounding vehicles are monitored. A hypothetical scenario is needed.

However, this virtual environment-based autonomous vehicle evaluation and verification focuses on vehicle performance and safety evaluation, and autonomous driving from the perspective of safe driving is required under the Road Traffic Act, which all road users, including autonomous drivers, must comply with while using public roads. There was a disappointment that it was not suitable for evaluating driving ability.

In addition, the current level of autonomous driving technology development is in progress for the commercialization of conditional autonomous driving, level 3, and advanced autonomous driving, level 4, according to the Society of Automotive Engineers (SAE) standards. For a considerable period of time after commercialization, autonomous vehicles will be They must drive in the same road space with other road users in mixed environmental situations. However, it is expected that the safety and smooth communication of road users due to autonomous vehicles will be limited. To this end, the rules that all road users must adhere to on the road are: The development of a traffic law-based evaluation system and scenarios for it needs to be concreted, but currently, there is almost no development of such road traffic law-based evaluation scenarios, so research and development on this is urgently needed.

Republic of Korea Patent No. 10-2158497 Republic of Korea Patent No. 10-2324978 Republic of Korea Patent No. 10-2368427 Republic of Korea Patent No. 10-2139513

The present invention was created to solve the conventional problems as described above. In order to evaluate the safe driving ability of self-driving cars based on the Road Traffic Act, which all road users must comply with, related laws are analyzed, and items and evaluations that self-driving cars must comply with are made. A self-driving car driving ability evaluation scenario based on road traffic laws that can improve evaluation reliability and evaluation accuracy by deriving elements and creating an evaluation scenario using an overseas standard scenario format as well as verifying whether the created evaluation scenario is logically error-free. The purpose is to provide a creation system.

However, the purpose of the present invention is not limited to this, and of course, even if not explicitly mentioned, purposes or effects that can be understood from the means of solving the problem or the embodiment are also included.

The road traffic law-based self-driving car driving ability evaluation scenario generation system of the present invention to achieve the above-mentioned purpose is a road traffic law-based self-driving car driving ability evaluation scenario generation system applied to a virtual environment-based self-driving car evaluation system, In order to obtain a vehicle operation permit, the autonomous vehicle's operating area, operating section, operating road information at the operating point, operating road signals, speed information, and precise map information provided by the appraised vehicle are received and identified based on the received information. , an information receiving unit that creates an operational design domain (ODD) through analysis and extraction; A single scenario generator that collects driving road information and performance data of the autonomous vehicle from the information receiver, reflects traffic law evaluation items, and converts them into ASAM OpenSCENARIO format to generate a single scenario; Based on the operation design area created by the information receiver, physical elements including road environment and facilities in the virtual environment are verified, and logical elements of vehicles operating on the road other than the evaluation vehicle are verified to determine whether the scenario can be performed. A scenario verification unit that determines and generates a verified verification scenario; A scenario operation DB management unit that collects and analyzes traffic law data applicable to the driving section from the information receiver and performs data management to create a complex scenario by layering the verification scenario data into two or more layers; It is characterized in that it includes a complex scenario generation unit that generates a complex scenario through the scenario operation DB management unit and transmits it to a platform for driving evaluation of an autonomous vehicle.

The information receiving unit receives operating road information, signal and speed information on operating roads, road sign information at points, and precise map information for obtaining permission to operate an autonomous vehicle, identifies the operating design area from the transmitted data, and analyzes it. By deriving and analyzing the basic environmental components for creating a virtual environment-based scenario, the operating road information, operating road signals, speed information, road sign information, precision map information, generated operating design area, and basic environmental components are derived and analyzed. It may be configured to transmit to a single scenario generator.

The single scenario generator defines the functional elements of the autonomous vehicle based on key performance data including acceleration, deceleration, stopping, and turning, which are the main functions of the autonomous vehicle, and uses the defined behavioral information to evaluate the autonomous vehicle. It reflects the driving elements including the road section, operating time, and route in the design area, and provides virtual environment-based evaluation section information based on the evaluation items applicable to a given driving section from the road traffic law evaluation items sent from the scenario operation DB management department. generates object information that reflects dynamic and static environmental elements within the section, creates a single scenario for law-based evaluation that can be performed on one road section based on the generated evaluation section information and object information, and ASAM OpenScenario format, it can be configured to generate a single scenario converted to ASAM OpenScenario format.

The scenario verification unit analyzes detailed values of ASAM format data, which is an overseas scenario standard, and verifies whether the road structure, weather and visibility conditions, and traffic safety facilities implemented in the virtual environment are implemented by comparing simulation values depicted in the actual vehicle. Perform a verification procedure to ensure that there are no logical errors in the evaluation criteria for road traffic law evaluation items, traffic volume, road users' object paths, and collisions that must be performed in the scenario, and determine whether the scenario can be performed by performing a simulation in a virtual environment. It may be configured to determine and generate verified verification scenario data.

The scenario operation DB management unit includes a Road Traffic Act evaluation item DB and an evaluation criteria DB that store and manage the driving section application law data transmitted from the information receiver, and the verification scenario data transmitted from the scenario verification unit is divided into evaluation types and roads. It can be configured to include a scenario DB that is stored and managed by stratifying by structure and driving section.

The composite scenario generator merges a single verification scenario that implements the driving section in a virtual environment from the scenario DB in accordance with the standards for legal compliance, and generates a composite scenario for evaluation of the merged scenario according to the standards in the evaluation standard DB. It can be configured to transmit the generated complex scenario to a virtual environment-based autonomous vehicle evaluation platform.

According to the above, the present invention improves the reliability and justification of the autonomous driving evaluation system based on the Road Traffic Act through the development of a scenario based on the Road Traffic Act for the construction and evaluation of an autonomous driving ability evaluation system for issuing autonomous driving licenses in preparation for commercialization of autonomous driving. It has the effect of securing.

In addition, the present invention builds a library of self-driving driving ability evaluation scenarios based on road traffic laws, so that self-driving industry workers, who are subject to prior evaluation, can utilize evaluation scenarios that may occur in situations where self-driving cars are mixed with regular vehicles to improve the level of self-driving technology. You can contribute.

In addition, the present invention has the effect of contributing to establishing a driver's license system exclusively for self-driving cars by utilizing future road traffic law-based self-driving car evaluation scenarios.

In addition, the present invention not only creates an evaluation scenario for self-driving cars based on the Road Traffic Law, but also verifies the generated evaluation scenario so that self-driving cars can drive safely while complying with the Road Traffic Law in road situations where other road users are mixed after the commercialization of autonomous driving. It can contribute to establishing driving ability evaluation standards and detailed judgment grounds for whether the driving ability can be performed.

In addition, the various and beneficial advantages and effects of the present invention are not limited to the above-described content, and may be more easily understood in the process of explaining specific embodiments of the present invention.

1 is a block diagram of a system for generating autonomous driving ability evaluation scenarios based on road traffic laws according to an embodiment of the present invention;
Figure 2 is an overall flowchart showing a method of generating an evaluation scenario by an autonomous driving ability evaluation scenario generation system based on road traffic laws according to an embodiment of the present invention;
Figures 3 to 7 are partial flow diagrams individually illustrating the detailed methods of each component in the overall flow diagram of Figure 2.

The above objects, other objects, features and advantages of the present invention will be easily understood through the following preferred embodiments related to the attached drawings. However, the present invention is not limited to the embodiments described herein and may be embodied in other forms. Rather, the embodiments introduced herein are provided so that the disclosed content will be thorough and complete and so that the spirit of the present invention can be sufficiently conveyed to those skilled in the art.

In this specification, when an element is referred to as being on another element, it means that it may be formed directly on the other element or that a third element may be interposed between them. Additionally, in the drawings, the thickness of components is exaggerated for effective explanation of technical content.

The terms used in this specification are for describing embodiments and are not intended to limit the invention. As used herein, singular forms also include plural forms, unless specifically stated otherwise in the context. As used in the specification, 'comprises' and/or 'comprising' does not exclude the presence or addition of one or more other elements.

In describing specific embodiments below, various specific details have been written to explain the invention in more detail and to aid understanding. However, a reader skilled in the art to understand the present invention will recognize that it can be used without these specific details. In some cases, it is mentioned in advance that when describing the invention, parts that are commonly known but are not significantly related to the invention are not described in order to prevent confusion without any reason in explaining the invention.

Hereinafter, with reference to the attached drawings, a road traffic law-based autonomous driving ability evaluation scenario generation system and evaluation scenario generation method according to an embodiment of the present invention will be described.

The road traffic law-based autonomous driving ability evaluation scenario generation system of the present invention is a road traffic law-based autonomous driving ability evaluation scenario that is the basis of a virtual environment-based autonomous vehicle evaluation system for the safety and smooth communication of road users in the autonomous driving era. It is a scenario creation system for creating.

Referring to FIG. 1, the road traffic law-based autonomous driving ability evaluation scenario generation system of the present invention includes an information reception unit 100, a single scenario generation unit 200, a scenario verification unit 300, and a scenario operation DB management unit 400. , and is configured to include a complex scenario creation unit 500.

Referring to Figures 1 to 3, the information receiver 100 receives driving road information, route information, and topographical information such as the driving area, driving section, and operating point of the autonomous vehicle provided by the person being evaluated for permission to operate the autonomous vehicle. , receives operating road signals, speed information, road surface sign information at points, and precise map information, and creates an operational design domain (ODD) by identifying, analyzing, and extracting the transmitted data (S110), The generated operation design area is analyzed to derive and analyze basic environmental components for creating a virtual environment-based scenario (S120).

The information receiving unit 100 includes the received operating road information such as the operating area, operating section, and operating point of the autonomous vehicle, route information, topographic information, signals of the operating road, speed information, road sign information at the point, and precise map information, It is configured to transmit the operation design area and basic environmental components created based on the received information to the single scenario generation unit 200.

Here, the appraisee refers to an autonomous vehicle manufacturer, a transportation company that operates autonomous vehicles such as self-driving buses, and self-driving taxis. When requesting permission to operate an autonomous vehicle through the evaluator's terminal, the operating area of the autonomous vehicle is determined. When driving road information such as driving sections and operating points, driving road signals, speed information, road sign information of points, and precise map information are transmitted, it can be configured to receive them from the information receiving unit 100.

Additionally, the information reception unit 100 may be configured to receive autonomous vehicle performance data provided by the person being evaluated and transmit the received performance data to the single scenario creation unit 200. However, self-driving car performance data is not limited to being provided by the evaluator, and may be configured to receive performance data measured for self-driving cars directly from an evaluation agency that evaluates self-driving cars. Here, the performance data may consist of performance information including acceleration, deceleration, stopping, turning, etc., which are the main functions of the autonomous vehicle.

Referring to FIGS. 1, 2, and 4, the single scenario creation unit 200 receives and collects driving road information and performance data of the autonomous vehicle from the information reception unit 100, and the scenario operation DB management unit 400 It can be configured to receive traffic law evaluation items from , reflect these traffic law evaluation items, and convert them into ASAM OpenSCENARIO format to create a single evaluation scenario.

The single scenario generation based on road traffic laws by the single scenario generator 200 may be configured to generate a scenario based on evaluation design elements such as evaluation scenario items, criteria, and difficulty level.

Specifically, the single scenario generator 200 defines autonomous vehicle functional elements based on performance data including acceleration, deceleration, stopping, turning, etc., which are the main functions of the autonomous vehicle (S210), and defines the defined behavioral information. It is configured to reflect autonomous vehicle driving elements, including road sections, operation times, routes, etc. in the operation design area for autonomous vehicle evaluation (S220).

Next, the single scenario generator 200 generates dynamic environmental elements (surrounding vehicles, PM (personal mobility), pedestrians, etc.) and static environmental elements (roads, traffic facilities, signs, etc.) within the section based on the autonomous vehicle driving elements. , etc.) is generated (S230), and a virtual road traffic law evaluation item applicable to the driving section is generated based on the road traffic law evaluation items provided by the road traffic law evaluation item DB (410) of the scenario operation DB management unit (400). It is configured to generate evaluation section information that reflects environment-based road traffic law evaluation items (S240).

Next, the single scenario generator 200 generates evaluation section information reflecting traffic law evaluation items, dynamic environmental elements (surrounding vehicles, PM (personal mobility), pedestrians, etc.) within the section, and static environmental elements (roads, traffic, etc.). Based on object information reflecting (facilities, signs, etc.), a single scenario for law-based evaluation that can be performed on one road section is created, converted to ASAM OpenScenario format, and a single scenario converted to ASAM OpenScenario format is created (S250) , It is configured to transmit a single scenario for law-based assessment converted to the ASAM OpenScenario format to the scenario verification unit 300. In addition, the single scenario generation unit 200 may be configured to transmit the operation design area received from the information reception unit 100 to the scenario verification unit 300.

Referring to FIGS. 1, 2, and 5, the scenario verification unit 300 verifies physical elements including the road environment and facilities in the virtual environment based on the operation design domain (ODD) (S310), and evaluates the evaluation vehicle and other After verifying the logical elements of the vehicle running on the road (S320), it is determined whether a single scenario converted to the ASAM OpenScenario format can be performed, and if it is verified to be possible, a verification single scenario is created (S330).

To be more specific, the scenario verification unit 300 adds detailed values of a single scenario for a law-based assessment composed of ASAM format data, which is an overseas scenario standard, to road structure, weather and visibility conditions, and traffic safety facilities implemented in a virtual environment. Verification is performed by comparing the simulation values depicted in the actual vehicle (S310), and the evaluation criteria for road traffic law evaluation items, traffic volume, and road users' object paths and collisions that must be performed in the scenario are logically error-free. Perform a verification procedure to determine whether the scenario can be performed (S320) and perform a simulation in a virtual environment to determine whether the scenario can be performed for the law-based evaluation single scenario transmitted from the single scenario generation unit 200 and generate verified verification single scenario data. It is configured (S330).

Referring to FIGS. 1, 2, and 6, the scenario operation DB management unit 400 is configured to include a Road Traffic Act evaluation item DB 410, an evaluation criteria DB 420, and a scenario DB 430.

The scenario operation DB management unit 400 receives the legal data applicable to the driving section from the information receiving unit 100, and stores the legal data applicable to the received driving section in the Road Traffic Act evaluation item DB 410 and the evaluation criteria DB 420. ) is configured to be stored and managed. In this case, when the driving section application law data is input to the information receiving unit 100, the information receiving unit 100 transmits the input driving section applicable law data to the scenario operation DB management unit 400, and the scenario operation DB management unit 400 It can be configured to receive the driving section applicable law data and store and manage it in a database in the Road Traffic Act evaluation item DB (410) and evaluation standard DB (420).

The scenario operation DB management unit 400 is configured to provide the road traffic law evaluation items included in the driving section application law data in the road traffic law evaluation item DB 410 to the single scenario generation unit 200.

The scenario operation DB management unit 400 is configured to store and manage the verification single scenario data transmitted from the scenario verification unit 300 by stratifying them by evaluation type, road structure, and driving section.

Referring to FIGS. 1, 2, and 7, the composite scenario creation unit 500 merges single scenarios from the scenario DB 400 and the scenario operation DB 430 within the Sanirio operation DB management unit 400 to create composite scenario data. It is configured to generate and transmit the generated complex scenario data to a virtual environment-based autonomous vehicle evaluation platform.

Specifically, the composite scenario generator 500 merges the verification single scenario in which the driving section is implemented in a virtual environment in the scenario DB 430 in accordance with the standards for traffic law compliance (S510), and uses the merged scenario as an evaluation standard. The DB 420 generates complex scenario data for driving evaluation according to the evaluation standard data (S520), and transmits the generated complex scenario data to a virtual environment-based autonomous vehicle evaluation platform.

Although the present invention has been shown and described in connection with preferred embodiments for illustrating the principles of the present invention, the present invention is not limited to the configuration and operation as shown and described. Rather, those skilled in the art will be able to understand that many changes and modifications can be made to the present invention without departing from the spirit and scope of the appended claims. Accordingly, all such appropriate changes, modifications and equivalents shall be considered to fall within the scope of the present invention.

100...Information receiving department
200...Single scenario generation unit
300...Scenario Verification Department
400...Scenario operation DB management department
410...Road Traffic Act Evaluation Item DB
420...Evaluation criteria DB
430...Scenario DB
500...complex scenario generation unit

Claims (6)

In the road traffic law-based autonomous driving ability evaluation scenario generation system applied to the virtual environment-based autonomous vehicle evaluation system,
In order to obtain permission to operate an autonomous vehicle, we receive and receive the operation area, operation section, operation road information of the operation point, signals of the operation road, speed information, and precise map information provided by the appraised person, and based on this received information, An information receiver that creates an operational design domain (ODD) through identification, analysis, and extraction;
A single scenario generator that collects driving road information and performance data of the autonomous vehicle from the information receiver, reflects traffic law evaluation items, and converts them into ASAM OpenSCENARIO format to generate a single scenario;
Based on the operation design area created by the information receiver, physical elements including road environment and facilities in the virtual environment are verified, and logical elements of vehicles operating on the road other than the evaluation vehicle are verified to determine whether the scenario can be performed. A scenario verification unit that determines and generates a verified verification scenario;
A scenario operation DB management unit that collects and analyzes traffic law data applicable to the driving section from the information receiver and performs data management to create a complex scenario by layering the verification scenario data into two or more layers;
It includes a complex scenario generation unit that generates a complex scenario through the scenario operation DB management unit and transmits it to a platform for driving evaluation of autonomous vehicles,
The single scenario generator,
Define self-driving car functional elements based on key performance data including acceleration, deceleration, stopping, and turning, which are the main functions of self-driving cars.
The defined behavioral information reflects driving elements including road sections, driving times, and routes in the driving design area for evaluating autonomous vehicles.
Virtual environment-based evaluation section information is created based on the evaluation items applicable to a given driving section from the road traffic law evaluation items sent from the scenario operation DB management department, and object information reflecting dynamic environmental elements and static environmental elements within the section is generated. ,
Based on the generated evaluation section information and object information, a single scenario for law-based evaluation that can be performed on one road section is created, converted to ASAM OpenScenario format, and road traffic law-based autonomy to generate a single scenario converted to ASAM OpenScenario format. Driving ability evaluation scenario generation system.
According to claim 1,
The information receiver,
Receive operating road information, signal and speed information on operating roads, road surface sign information at points, and precision map information to obtain permission to operate an autonomous vehicle, and identify and analyze the operating design area from the transmitted data, based on a virtual environment. By deriving and analyzing the basic environmental components for scenario creation, the driving road information, driving road signals, speed information, road surface sign information, precision map information, generated driving design area, and basic environmental components are stored in a single scenario creation unit. A road traffic law-based autonomous driving ability evaluation scenario generation system characterized by transmission.
delete According to claim 1,
The scenario verification unit,
By analyzing the detailed values of the ASAM format data, which is an international scenario standard, we verify whether the road structure, weather and visibility conditions, and traffic safety facilities implemented in the virtual environment are implemented by comparing the simulation values depicted in the actual vehicle.
We carry out verification procedures to ensure that there are no logical errors in the evaluation criteria for road traffic law evaluation items, traffic volume, road users' object paths, and collisions that must be performed in the scenario, and determine and verify whether the scenario can be performed by performing a simulation in a virtual environment. A road traffic law-based autonomous driving ability evaluation scenario generation system characterized by generating verification scenario data.
According to claim 1,
The scenario operation DB management department
Contains a Road Traffic Act evaluation item DB and an evaluation criteria DB that store and manage the driving section applicable law data transmitted from the information receiver,
A road traffic law-based autonomous driving ability evaluation scenario generation system comprising a scenario DB that stores and manages the verification scenario data transmitted from the scenario verification unit by stratifying them by evaluation type, road structure, and driving section.
According to claim 5,
The complex scenario creation unit
In the scenario DB, a single verification scenario that implements the driving section in a virtual environment is merged in accordance with the standards for legal compliance, and a composite scenario for evaluation is created in accordance with the standards in the evaluation standard DB for the merged scenario. A road traffic law-based autonomous driving ability evaluation scenario generation system that is configured to transmit the scenario to a virtual environment-based autonomous vehicle evaluation platform.

Images