A legal safety concept for highly automated driving on highways

Vanholme, Benoit; Gruyer, Dominique; Glaser, Sebastien; Mammar, Said · 2011 · Crossref

DOI: 10.1109/ivs.2011.5940582

archive: archived pipeline: cataloged verified

Get this paper ↗ (DOI — opens at the source; we link to it, we don't host it)

Summary

This paper proposes a "legal safety" concept for Highly Automated Driving Systems (HAS) on highways, designed to ensure safety when all road users strictly adhere to traffic rules. Motivated by the need for automated vehicles to share infrastructure with human drivers during a transitional period, the authors argue that traffic laws, specifically the 1968 Vienna Convention on Road Traffic, can serve as the foundational framework for system design. The concept assumes that if the automated system and other road users respect legal norms, safety is guaranteed; if a human driver violates rules, the system attempts to avoid accidents or performs emergency braking. The study focuses on a HAS with speed keeping, distance keeping, and lane changing functionalities. The system architecture comprises perception, co-pilot (decision-making), and control modules. The perception module uses sensors to detect lanes, traffic signs, and objects, while the co-pilot calculates safe trajectories using a curvilinear lane coordinate system. The authors derive specific requirements from Vienna Convention articles, such as maintaining safe distances (Art. 5), overtaking only on the left (Art. 4), and adapting speed to conditions (Art. 5). The co-pilot predicts object trajectories by assuming legal behavior but adopts a defensive stance by considering worst-case scenarios, including "phantom" objects at the edge of the perception horizon to handle unknown obstacles. The core contribution is the mathematical formulation of trajectory generation and speed profiles. The co-pilot generates ten potential trajectories: six for normal operation (one per lane with two speed profiles) and four for emergency states. Heuristics limit the solution space to ensure computational efficiency. The system calculates target positions and speeds based on safety distances proportional to object speeds and extreme deceleration capabilities. It ensures that the ego vehicle can always stop for a traffic jam or avoid collisions with phantoms. The design also accounts for uncertainty in object behavior, such as potential lane changes, by predicting multiple trajectory variants. The significance of this work lies in providing a rigorous, rule-based framework for automated driving that facilitates coexistence with human drivers. By grounding the system in existing international traffic laws, the authors offer a legally defensible and socially acceptable approach to automation. The paper demonstrates that a highly automated system can be designed to be both safe and compliant with traffic regulations, addressing key challenges in perception, prediction, and control. This approach supports the incremental introduction of ADAS and provides a basis for future research in mixed-traffic environments.

Provenance

The full processing record for this entry. Every stage of this paper's journey through the pipeline is logged — what ran, with which tool and model, how many attempts it took, and when it last completed.

Summary generated by qwen3.6-27b-nvidia on 2026-08-10; verification: verified.

Topics

Ranked by relevance to this paper. Hover a topic for its definition.

• situational awareness
• driverless ads

Information type

What kind of knowledge this paper contributes, grouped by family — independent of topic (what it is about) and method (how it was studied).

• Theoretical Contribution: computational model, conceptual framework