When will automated vehicles come onto the market? This question has puzzled the automotive industry and society for years. The technology and its implementation have made rapid progress over the last decade, but the challenge of how to prove the safety of these systems has not yet been solved. Since a market launch without proof of safety would neither be accepted by society nor by legislators, much time and many resources have been invested into safety assessment in recent years in order to develop new approaches for an efficient assessment. This paper therefore provides an overview of various approaches, and gives a comprehensive survey of the so-called scenario-based approach. The scenario-based approach is a promising method, in which individual traffic situations are typically tested by means of virtual simulation. Since an infinite number of different scenarios can theoretically occur in real-world traffic, even the scenario-based approach leaves the question unanswered as to how to break these down into a finite set of scenarios, and find those which are representative in order to render testing more manageable. This paper provides a comprehensive literature review of related safety-assessment publications that deal precisely with this question. Therefore, this paper develops a novel taxonomy for the scenario-based approach, and classifies all literature sources. Based on this, the existing methods will be compared with each other and, as one conclusion, the alternative concept of formal verification will be combined with the scenario-based approach. Finally, future research priorities are derived.

/pdf/survey-on-scenario-based-safety-assessment-of-automated-2u2a6wl7tk.pdf

Survey on Scenario-Based Safety Assessment of Automated Vehicles

Simulations are commonly used to validate the design of autonomous systems. However, as these systems are increasingly deployed into safety-critical environments with aleatoric uncertainties, and with the increase in components that employ machine learning algorithms with epistemic uncertainties, validation methods which consider uncertainties are lacking. We present an approach that evaluates signal propagation in logical system architectures, in particular environment perception-chains, focusing on effects of uncertainty to determine functional limitations. The perception based autonomous driving systems are represented by connected elements to constitute a certain functionality. The elements are based on (meta-)models to describe technical components and their behavior. The surrounding environment, in which the system is deployed, is modeled by parameters that are derived from a quasi-static scene. All parameter variations completely define input-states for the designed perception architecture. The input-states are treated as random variables inside the model of components to simulate aleatoric/epistemic uncertainty. The dissimilarity between the model-input and -output serves as measure for total uncertainty present in the system. The uncertainties are propagated through consecutive components and calculated by the same manner. The final result consists of input-states which model uncertainty effects for the specified functionality and therefore highlight shortcomings of the designed architecture.

A Systematic Approach to Analyzing Perception Architectures in Autonomous Vehicles

Scenarios are the basis of virtual simulation and should be properly designed and created to ensure the diversity, rationality and criticality. Many researches involve the description of scenario and generation of test case. In this work, we propose a new SOTIF scenario hierarchy to depict static entities, dynamic information and relationships in the scenario. The concept of scenario potential risk is defined in this paper. The prior collision probability and safety control intensity are introduced to quantize the scenario potential risk. Finally, based on the architecture of scenario, we designed a novel process to automatically generate scenarios with the genetic algorithm. The experiments carried out show that proposed methods can generate diverse and trivial test case library.

A new SOTIF scenario hierarchy and its critical test case generation based on potential risk assessment

Ontology is the explicit and formal representation of the concepts in a domain and relations among them. Transportation science is a wide domain dealing with mobility over various complex and interconnected transportation systems, such as land, aviation, and maritime transport, and can take considerable advantage from ontology development. While several studies can be found in the recent literature, there exists a large potential to improve and develop a comprehensive smart mobility ontology. The current chapter aims to present different aspects of ontology development in general, such as ontology development methods, languages, tools, and software. Subsequently, it presents the currently available mobility-related ontologies developed across different domains, such as transportation, smart cities, goods mobility, sensors. Current gaps in the available ontologies are identified, and future directions regarding ontology development are proposed that can incorporate the forthcoming autonomous and connected vehicles, mobility as a service (MaaS), and other disruptive transportation technologies and services.

Smart Mobility Ontology: Current Trends and Future Directions.

Motion planning is a major component of any automated driving system. The safety assessment of such components requires a formal characterization of the perception and control mechanisms. This requires dedicated tools and models for the environment, sensors and vehicles that are highly representative of the real world. Simulation is a method to virtually investigate the behavior of systems under study. It has a key role to play in demonstrating the safety of autonomous vehicles. In this context, we consider a control module as a black-box and try to determine a reference which represents the ’right decision’, if it exists. An optimization-based reference model is created for the control function. This model allows each scene in the environment to be mapped to the desired decision regardless of the black-box. The black-box and the reference model are run on several critical scenarios. In output, an assessment of decision making is performed along with systematic criticality characterization of targeted scenarios.

/pdf/a-simulation-based-framework-for-functional-testing-of-3ia0thsocc.pdf

A simulation-based framework for functional testing of automated driving controllers

Autonomous vehicles perceive the environment with different kinds of sensors (camera, radar, lidar...). They must evolve in an unpredictable environment and a wide context of dynamic execution, with strong interactions. In order to generate the safety of the autonomous vehicle, its occupants and the others road users, it is necessary to validate the decisions of the algorithms for all the situations that will be met. These situations are described and generated as different use cases of automated vehicles. In this work, we propose an approach to generate automatically use cases of autonomous vehicle for highway. This approach is based on a three layers hierarchy, which exploits static and mobile concepts we have defined in the context of three ontologies: highway, weather and vehicle. The highway ontology and the weather ontology conceptualize the environment in which evolves the autonomous vehicle, and the vehicle ontology consists of the vehicle devices and the control actions. To apply our approach, we consider a running example about the insertion of a vehicle by the right entrance lane of a highway.

https://hal.archives-ouvertes.fr/hal-01835139/document

An Ontology-based Approach to Generate the Advanced Driver Assistance Use Cases of Highway Traffic

Autonomous vehicles perceive the environment with different kinds of sensors (camera, radar, lidar...). They must evolve in an unpredictable environment and a wide context of dynamic execution, with strong interactions. Therefore, ensuring the functionality and safety of the autonomous driving system has become one of the focuses of research in the field. In order to guarantee the safety of the autonomous vehicle, its occupants and the others road users, it is necessary to validate the decisions of the algorithms for all the situations that will be met by the vehicle. These situations are described and generated as different scenarios. The main objective of this work is to generate all these scenarios and find out the critical ones. Therefore, we use a scenario-generation methodology which uses the Performance Evaluation Process Algebra (PEPA) for modelling the transitions between the driving scenes. To apply our approach, we consider a running example about a riding autonomous vehicle in the context of a three-lane highway.

Stochastic Modelling of Autonomous Vehicles Driving Scenarios Using PEPA

Autonomous cars mainly rely on an intelligent system pilot to achieve the purpose of self-driving They combine a variety of sensors to perceive their surroundings, such as cameras, radars and lidars The perception algorithms of the Advanced Driver-Assistance Systems (ADAS) provide observations on the environmental elements based on the data provided by the sensors, while decision algorithms generate the actions to be implemented by these vehicles To ensure the safety of the autonomous vehicle, it is necessary to specify, validate and secure the dependability of the architecture and the behavioural logic of ADAS running on the vehicle for all the situations that will be met by the vehicle These situations are described and generated as different test cases In this work, we propose a methodology to generate automatically test cases of autonomous vehicle for highway This methodology is based on a three layers hierarchy The first layer exploits static and mobile concepts we have defined in the context of three ontologies: highway, weather and vehicle The second layer exploits the relationships between these concepts while the third one exploits the method of test case generation based on the first two layers Finally, we use the Performance Evaluation Process Algebra (PEPA) for modelling the transitions between the driving scenes To apply our methodology, we consider a running example about a riding vehicle on the left of the autonomous vehicle to take a right exit lane of a highway

Wei Chen

Papers

An Ontology-based Approach to Generate the Advanced Driver Assistance Use Cases of Highway Traffic

Stochastic Modelling of Autonomous Vehicles Driving Scenarios Using PEPA

An Advanced Driver Assistance Test Cases Generation Methodology Based on Highway Traffic Situation Description Ontologies.