Highly automated driving on highways: System implementation on PC and automotive ECUs

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

DOI: 10.1109/itsc.2011.6083143

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 addresses the implementation of a highly automated driving system for highways, aiming to reduce accidents caused by human error such as distraction and fatigue. Motivated by the need for safer, more comfortable transport, the authors integrate components from two European research projects: the HAVEit project, which provided the co-pilot and control algorithms, and the ABV project, which contributed perception technologies. The system is designed to collaborate with a human driver rather than replace them entirely, utilizing a step-by-step introduction of automation. The system architecture comprises four main components: perception, co-pilot, control, and human-machine interface (HMI) with a mode selection unit (MSU). The perception module fuses data from cameras, laser scanners, radars, and infrastructure-to-vehicle communication to map lanes, objects, and traffic signs in a vehicle-fixed coordinate system. The co-pilot calculates optimal trajectories based on a "legal safety" concept, predicting object behaviors and ensuring compliance with traffic rules. The control component guides the vehicle along these trajectories. Implementation was split between hardware and software: safety-critical co-pilot and control algorithms were deployed on AUTOSAR-based automotive Electronic Control Units (ECUs), while perception, HMI, and MSU components ran on a standard PC using the RTMaps environment. Development followed a V-cycle methodology, utilizing the SiVIC simulation tool for software-in-the-loop testing before transferring the system to a physical vehicle for hardware-in-the-loop validation on a test track. Experimental results demonstrate the system's capability to manage highway scenarios. In a speed limit approach scenario, the vehicle successfully detected a future 30 km/h limit while traveling at 50 km/h and adapted its speed profile to reach the limit with minimal braking, maintaining a control error of 1.5 km/h. In an overtaking scenario, the system correctly interpreted an object’s indicator signals, prioritized the object’s lane change, and subsequently proposed a safe lane change to overtake once the path was clear. The co-pilot and control algorithms on the ECUs operated with cycle times below 25 ms. The HMI provided visual feedback on optimal speed and lane, while haptic interfaces allowed for intuitive driver interaction. The significance of this work lies in demonstrating the feasibility of integrating complex automated driving functions onto automotive-grade ECUs while maintaining real-time performance. The study validates a modular architecture that separates safety-critical decision-making and control from perception and interface management. The authors conclude that current computational loads allow for the potential consolidation of all algorithms onto a single ECU in future developments. Future work will focus on testing at higher speeds, implementing dynamic lane changes on physical vehicles, and conducting user acceptance studies to refine the HMI and system parameters.

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
• adaptive driving beam

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).

• Methodological Resource: tool software
• Theoretical Contribution: computational model