Autonomous vehicle control at the limits of handling

Kritayakirana, Krisada; Gerdes, J. Christian · 2012 · OpenAlex-citations

DOI: 10.1504/ijvas.2012.051270

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 challenge of controlling autonomous vehicles at the limits of tire adhesion, a capability possessed by expert racecar drivers that could significantly enhance vehicle safety systems. The authors argue that previous approaches, which often couple path generation and control into single optimization problems, lack physical intuition and robustness against real-world friction variations. To bridge this gap, the study designs an autonomous racing controller that mimics the decision-making processes of racecar drivers, specifically separating path planning from vehicle control to provide insights for future driver assistance technologies. The methodology involves constructing a controller architecture based on a quasi-static bicycle model and a ‘g-g’ diagram, which represents the vehicle’s friction limits. Path generation utilizes clothoid curves to create smooth transitions between straight sections and constant-radius corners, mimicking a racing line. The control system comprises feedforward and feedback components. Feedforward inputs for steering and longitudinal acceleration are calculated using the bicycle model and the friction limit circle to ensure the vehicle operates at its adhesion limits. Feedback control imitates driver corrections through three modules: a lanekeeping steering feedback system to minimize tracking errors, a yaw damping term to reduce oscillations, and a slip circle-based longitudinal controller that modulates throttle and brake inputs to regulate wheel slip and maintain stability. Experimental validation was conducted on an Audi TTS equipped with instrumentation, operating on a low-friction surface with all electronic driving aids, such as Electronic Stability Program and Anti-lock Braking System, disabled. The results demonstrate that the controller can robustly track a desired path while operating at the limits of tire adhesion. The lanekeeping and wheel slip feedback controllers effectively maintained stability and tracking accuracy, even in scenarios where friction was overestimated. The experiments highlighted the critical importance of coordinating steering and longitudinal inputs, confirming that the modular controller design successfully replicated the complex dynamics of limit handling. The significance of this work lies in its demonstration that autonomous systems can achieve racecar-level handling capabilities through a physically intuitive, modular control structure. By separating path planning from control and using feedback mechanisms that mimic human driver responses, the controller provides a robust framework for operating at friction limits. This approach not only advances autonomous racing but also offers a scalable foundation for driver assistance systems, where individual control modules can be deployed to assist human drivers in maintaining vehicle stability during extreme maneuvers.

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-prismaquant on 2026-06-26; verification: verified.

Topics

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

• steering pattern

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