A hybrid model predictive control for traffic flow stabilization and pollution reduction of freeways

Csikós, Alfréd; Varga, István; Hangos, Katalin M. · 2018 · Crossref

DOI: 10.1016/j.trd.2018.01.006

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 dual challenge of stabilizing freeway traffic flow and reducing pollutant concentrations in inhabited areas adjacent to motorways. The authors identify a conflict between conventional traffic stabilization measures, such as variable speed limits (VSL), and pollution reduction objectives, as VSL can increase traffic density and subsequently raise emissions. To resolve this, the study proposes a hybrid model predictive control (NMPC) system that separates control tasks into two distinct modes: one for maintaining pollutant levels below legislative limits under stable traffic conditions, and another for suppressing shockwaves during unstable conditions. The methodology integrates the second-order macroscopic traffic model METANET with a previously developed emission dispersion model. The traffic model describes vehicle density, mean speed, and ramp queues across discrete freeway segments, while the dispersion model calculates pollutant concentrations (CO, HC, NOx) in balance volumes near the motorway, accounting for wind direction and speed. The joint system is controlled via ramp metering and VSL inputs. The hybrid controller structure is realized using a finite automata that switches between control modes based on predefined stability rules. In the first mode, only ramp metering is utilized to minimize ramp queues while keeping concentrations under specified limits. In the second mode, triggered by shockwave threats, both ramp metering and VSL are employed to stabilize traffic, temporarily neglecting concentration constraints. The paper details the mathematical formulation of the joint state-space model, including dynamic equations for traffic flow and pollutant dispersion. It specifies the control inputs, disturbances (such as upstream flow and wind speed), and measured outputs. The control design employs nonlinear model predictive control techniques with distinct cost functions for each mode to optimize performance. A complex case study is presented to evaluate the controller, analyzing the performance of individual control modes and the switching behavior between them. The significance of this work lies in its integration of macroscopic traffic dynamics with spatiotemporal emission dispersion models within a unified control framework. By separating control objectives into a hybrid structure, the system avoids the suboptimal solutions often resulting from multi-objective designs that treat pollution as a soft constraint. This approach ensures that traffic stabilization measures do not inadvertently exacerbate pollution levels during stable conditions, while still providing robust shockwave suppression when necessary. The study demonstrates that dedicated controllers for specific operational modes, managed by a stability-preserving switching mechanism, offer a more effective strategy for simultaneous traffic management and environmental protection.

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.

• traffic density