A Comprehensive Taxonomy for Classifying Automated Driving Systems Based on Operational Design Domain, Automation Level, and Technological Readiness

Core Concepts

This paper proposes a new taxonomy for automated driving systems that combines the operational design domain, level of automation, and technological readiness to provide a more comprehensive and structured approach for categorizing and comparing these systems.

Abstract

The paper presents a three-step methodology to develop a new taxonomy for automated driving systems (ADS):

Structuring Operational Design Domains (ODD) with an intermediate-level taxonomy: The authors define five key categories (country code, road users, road types, environmental conditions, and velocity) with predefined attributes to describe the ODD of an ADS in a concise manner.

Adding Levels of Automation: The authors incorporate the well-known SAE Levels 0-5 to capture the degree of automation and the driver's responsibility when the automated driving function is activated.

Estimating AD-Readiness: The authors propose a new "Automated Driving-Readiness Level" (ADRL) model, inspired by the Technology Readiness Level (TRL) framework, to assess the maturity of the ADS technology.

The authors then apply the proposed taxonomy to several examples of current and future ADS, including truck highway pilots, valet parking systems, highway pilots, robotaxis, and mining trucks. This demonstrates the applicability of the new taxonomy and highlights the potential to identify "white spots" in the development and regulation of ADS.
The authors acknowledge that the taxonomy may need to be refined as new systems and developments emerge, but they believe the proposed approach provides a more comprehensive and structured way to categorize and compare ADS, which can benefit both industry and regulators.

Stats

None.

Quotes

None.

Key Insights Distilled From

A new Taxonomy for Automated Driving: Structuring Applications based on their Operational Design Domain, Level of Automation and Automation Readiness

by Johannes Bet... at arxiv.org 04-29-2024

https://arxiv.org/pdf/2404.17044.pdf

Deeper Inquiries

How can the proposed taxonomy be extended to capture the potential for collaboration and cooperation between autonomous vehicles and infrastructure?

The proposed taxonomy can be extended to incorporate the potential for collaboration and cooperation between autonomous vehicles and infrastructure by introducing new categories and attributes that specifically address these aspects. One way to achieve this is by including categories related to communication protocols, data sharing mechanisms, and infrastructure requirements. Attributes could focus on the ability of autonomous vehicles to interact with smart infrastructure, such as traffic lights, road signs, and other vehicles, to optimize traffic flow and enhance safety.
Moreover, the taxonomy could include categories for vehicle-to-infrastructure (V2I) and vehicle-to-vehicle (V2V) communication capabilities, highlighting the level of connectivity and coordination between different entities. Attributes within these categories could cover aspects like real-time data exchange, cooperative decision-making algorithms, and shared situational awareness.
By integrating these elements into the taxonomy, it would provide a more comprehensive framework for assessing the readiness of autonomous systems to collaborate with infrastructure elements, paving the way for more efficient and intelligent transportation systems.

How might the proposed taxonomy be integrated with existing safety frameworks and regulatory processes to facilitate the deployment of automated driving systems?

Integrating the proposed taxonomy with existing safety frameworks and regulatory processes is crucial for ensuring the safe and effective deployment of automated driving systems. One way to achieve this integration is by aligning the taxonomy categories and attributes with established safety standards and guidelines, such as ISO 26262 for functional safety in automotive systems.
Additionally, the taxonomy can be linked to regulatory requirements specific to automated driving, such as UN regulations and national legislation governing autonomous vehicles' operation. By mapping the taxonomy elements to safety-critical functions and performance metrics outlined in these regulations, stakeholders can assess compliance and identify areas for improvement.
Furthermore, the taxonomy can serve as a tool for risk assessment and mitigation, allowing regulators to evaluate the safety implications of different automation levels and operational design domains. By incorporating safety-related attributes into the taxonomy, such as fail-safe mechanisms, emergency response protocols, and cybersecurity measures, it can provide a comprehensive framework for evaluating the safety readiness of automated driving systems.
Overall, by integrating the proposed taxonomy with existing safety frameworks and regulatory processes, stakeholders can streamline the approval process, enhance transparency, and ensure that automated driving systems meet the necessary safety standards before deployment.

What are the key challenges in developing a universally accepted taxonomy for automated driving systems that can accommodate the rapid technological advancements in this field?

Developing a universally accepted taxonomy for automated driving systems that can keep pace with rapid technological advancements poses several challenges:

Technological Complexity: The evolving nature of autonomous vehicle technology, including AI algorithms, sensor systems, and connectivity solutions, makes it challenging to create a taxonomy that captures all relevant aspects without becoming overly complex.

Interdisciplinary Nature: Automated driving systems involve a wide range of disciplines, from engineering and computer science to law and ethics. Developing a taxonomy that accommodates these diverse perspectives and ensures consensus among stakeholders is a significant challenge.

Standardization: Achieving standardization across different regions and industries is crucial for a universally accepted taxonomy. Harmonizing terminology, definitions, and classification criteria can be complex, especially in a global context.

Dynamic Environment: The rapid pace of technological advancements in the field of automated driving requires a taxonomy that is flexible and adaptable to new innovations and emerging use cases. Keeping the taxonomy up-to-date with the latest developments is a continuous challenge.

Regulatory Compliance: Ensuring that the taxonomy aligns with existing regulations and safety standards while also accommodating future regulatory changes is essential but challenging. Balancing innovation with compliance is a delicate task.

User Understanding: Making the taxonomy accessible and understandable to a wide range of users, including policymakers, industry professionals, and the general public, is crucial for its acceptance and adoption. Simplifying complex technical concepts without oversimplifying them is a key challenge.

Addressing these challenges requires collaboration among experts from various fields, ongoing stakeholder engagement, regular updates to the taxonomy, and a commitment to transparency and inclusivity in the development process.