Formulation and interpretation of optimal braking and steering patterns towards autonomous safety-critical manoeuvres

Fors, Victor; Olofsson, Björn; Nielsen, Lars · 2018 · OpenAlex-citations

DOI: 10.1080/00423114.2018.1549331

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 optimizing vehicle stability control for autonomous safety-critical maneuvers, specifically shifting the focus from traditional yaw control to lane-keeping or lane-changing objectives. The authors aim to exploit the maximum available tire forces through optimal braking and steering inputs to enhance vehicle safety and autonomy. The research is motivated by the need for active-safety systems that can adapt braking intensity to specific situations, balancing the trade-off between maximizing initial velocity (requiring heavy braking to stay in lane) and maximizing final velocity (requiring less braking to maintain speed). To achieve this, the authors formulate an optimal control problem (OCP) using a double-track vehicle model that includes longitudinal and lateral load transfer, nonlinear tire forces via Pacejka’s Magic Formula, and wheel dynamics. The optimization criterion is a linear combination of the initial velocity ($v_0$) and final velocity ($v_f$), weighted by an interpolation parameter $\eta$ ranging from 0 to 1. The study evaluates three scenarios: a left-hand turn, a double lane-change based on ISO standard 3888-2, and an avoidance maneuver. The OCP is solved numerically using direct collocation and the IPOPT solver, with constraints on wheel torques, steering rate, and lane boundaries. The results reveal a continuous family of optimal braking and steering patterns dependent on $\eta$. When $\eta = 1$, the strategy prioritizes maximizing initial velocity, resulting in heavy all-wheel braking to reduce speed and maintain lane position, with tire forces operating near the friction ellipse limit. Conversely, when $\eta = 0$, the strategy prioritizes final velocity, resembling traditional yaw control where braking is applied primarily to inner wheels to generate yaw moments without significant speed reduction. Intermediate values of $\eta$ produce hybrid strategies that combine turn-in and turn-out moments. In the left-hand turn, higher $\eta$ values lead to greater body slip and curvature, while lower $\eta$ values sacrifice some tire-force utilization for better lateral force availability and reduced body slip. Similar patterns emerge in the double lane-change and avoidance scenarios, where the presence of future obstacles influences the braking strategy. The significance of this work lies in providing a unified framework that embeds both optimal lane-keeping and traditional yaw control as boundary cases of a single optimization criterion. This offers new insights for designing adaptive safety systems that can modulate braking aggressiveness based on situational margins. For instance, less braking-intensive strategies can reduce the risk of rear-end collisions caused by sudden heavy braking or allow for earlier intervention with a safety margin. The findings suggest that autonomous vehicles can exploit these invariant strategies to optimize performance across various critical maneuvers, improving both safety and handling efficiency.

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