Performance assessment of an adaptive model predictive control with torque braking for lane changes

Zulkarnain, Zulkarnain; Bizzy, Irwin; Sofijan, Armin; Ariff, Mohd Hatta Mohammed · 2026 · Crossref

DOI: 10.12928/telkomnika.v24i2.27167

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 maintaining safety and stability in autonomous vehicles during complex maneuvers, specifically lane changes, under diverse and unpredictable driving conditions. Existing control systems often struggle to manage the combined demands of steering and braking when faced with varying speeds and low-friction road surfaces. To bridge this gap, the authors propose and evaluate an adaptive model predictive control (MPC) system integrated with a torque braking distribution strategy. The primary motivation is to develop a robust controller that can continuously update its internal model in real-time, allowing it to anticipate and respond to changing vehicle dynamics and road friction more effectively than static controllers. The study employs a simulation-based experimental design using a high-fidelity, non-holonomic vehicle dynamics model. The control framework utilizes a discrete-time state-space model for the vehicle plant, incorporating nonlinear tire forces via the Pacejka tire model. The adaptive MPC controller optimizes steering angle and yaw moment inputs by minimizing a quadratic cost function that balances tracking accuracy and control effort. A key component of the method is a recursive least squares (RLS) estimator that updates critical parameters, such as the tire-road friction coefficient, online. Additionally, a torque braking distribution strategy calculates specific braking torques for individual wheels to generate stabilizing yaw moments. The system was tested across speeds of 10–25 m/s and road friction coefficients ranging from 0.3 (icy) to 1.0 (dry asphalt) using double-lane-change trajectories. The results demonstrate significant performance improvements for the proposed integrated controller compared to a non-adaptive MPC baseline. The adaptive MPC achieved a 52.8% average reduction in lateral tracking error and reduced yaw rate by up to 41.8% on low-friction surfaces. Specific simulations at 15 m/s on a surface with a friction coefficient of 0.6 showed that the torque braking distribution proactively corrected the vehicle’s path, with differential torque generation between front wheels creating necessary stabilizing yaw moments. However, root mean square error analysis revealed performance limitations at high speeds (>22.5 m/s) on low-friction roads, where errors increased sharply due to tire force saturation and nonlinear dynamics that the controller could not fully compensate for. The significance of this work lies in its validation of synergistic steering and braking coordination within a unified predictive optimization loop. The findings confirm that real-time parameter adaptation and proactive torque vectoring significantly enhance the precision and reliability of autonomous lane changes. While the study is limited to simulation, it provides quantitative evidence that integrated chassis control strategies are critical for maintaining stability under adverse conditions. The authors conclude that future work must address model uncertainties through hardware-in-the-loop testing and prototype vehicle experiments to validate real-world efficacy.

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.

• lane changing