Model-based Predictive Control implementation for Cooperative Adaptive Cruise Control

Lopes, António; Araújo, Rui Esteves · 2018 · Crossref

DOI: 10.24840/2183-6493_002.001_0001

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 need for robust automated vehicle control systems to mitigate traffic congestion, energy consumption, and safety risks associated with human error. Specifically, it focuses on Cooperative Adaptive Cruise Control (CACC), a paradigm that utilizes vehicle-to-vehicle communication to access state information from preceding vehicles, allowing for tighter, safer platooning than traditional sensor-based adaptive cruise control. The authors propose a Model Predictive Control (MPC) implementation to manage longitudinal velocity and spacing distance in a two-vehicle platoon, aiming to ensure system stability and collision avoidance. The methodology involves a two-layer control structure. The lower layer employs feedback linearization to handle nonlinear vehicle dynamics, reducing the system to a linear model where vehicle jerk serves as the control input. The upper layer utilizes MPC to optimize this jerk input. The system is modeled using discrete-time state-space equations derived via forward Euler approximation, with state variables including relative distance, relative velocity, and relative acceleration. The MPC design computes future state predictions over a finite horizon and determines the optimal control sequence by minimizing a quadratic cost function. This optimization incorporates hard constraints to limit control rates (jerk) for comfort and ensures the relative distance remains positive to prevent collisions. The proposed controller was evaluated through Simulink simulations involving two identical vehicles. The lead vehicle maintained a constant velocity of 20 m/s, while the controlled follower vehicle started at 18 m/s with an initial relative distance of 10 meters. The controller successfully adjusted the follower’s velocity to match the leader and reduced the spacing to 1 meter. The study analyzed the impact of varying the gain matrix $G$ and control rate limits. Results indicated that increasing the gain $G$ slowed the system response but produced smoother control variables. Conversely, imposing stricter limits on the control rate resulted in slower responses and overshoot, while looser limits allowed for more abrupt control actions. The constraint ensuring positive relative distance effectively prevented the system from entering unsafe states. The study concludes that the MPC approach effectively balances control responsiveness with safety and comfort constraints. By integrating prediction models with optimization algorithms, the controller can maintain safe following distances and velocity synchronization. The findings demonstrate that tuning gain matrices and control limits allows for trade-offs between response speed and smoothness. The authors suggest future work should expand the model to include larger platoons, incorporate additional constraints, and analyze the string stability of the system.

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.

• following distance