A Cooperative Longitudinal-Lateral Platoon Control Framework with Dynamic Lane Management for Unmanned Ground Vehicles Based on a Dual-Stage Multi-Objective MPC Approach

Shunchao Wang, Zhigang Wu, Yonghui SuDrones 2025 · DOI

TL;DR

This paper introduces a novel cooperative control framework for unmanned ground vehicle (UGV) platoons that tackles challenges in mixed traffic. It uses a dual-stage, multi-objective Model Predictive Control (MPC) approach with adaptive weighting to balance platoon efficiency and overall traffic flow, alongside a dynamic lane management system to optimize UGV-dedicated lane usage. This leads to significantly faster platoon formation with minimal impact on traffic speed.

What Problem Does This Paper Address?

Forming stable and efficient platoons of autonomous vehicles (UGVs) is crucial for improving road capacity and safety. However, existing methods struggle with:

  • Low UGV penetration: When there aren't many UGVs, it's hard to form platoons quickly, limiting efficiency gains.
  • Fragmented control: Many approaches treat longitudinal (speed) and lateral (lane changing) control separately, leading to suboptimal coordination.
  • Rigid infrastructure: Dedicated UGV lanes can be underutilized, especially at low penetration rates, wasting road capacity.
  • Mixed traffic uncertainty: The unpredictable behavior of human-driven vehicles (HDVs) complicates platoon formation and stability.

This paper aims to create a unified, adaptive framework that addresses these issues, enabling robust UGV platooning in dynamic, mixed-traffic environments.

Methods

The core of the proposed solution is a hierarchical cooperative control architecture for UGV platooning. Key components include:

  • Dual-Stage Multi-Objective Model Predictive Control (MPC): This is the central control strategy. It breaks down the complex task into two phases:
  • Pursuit Phase: The following UGV approaches the leading UGV.
  • Platooning Phase: The UGVs establish and maintain a stable platoon formation. Each phase uses MPC to optimize control actions over a short horizon, considering vehicle dynamics and traffic interactions.

  • Adaptive Weighting Strategy: The multi-objective MPC has several competing goals (e.g., fast formation, smooth traffic, efficient speed). This strategy dynamically adjusts the importance (weights) of these objectives based on the current traffic conditions (e.g., congestion, spacing) and the phase of platoon formation. This allows the system to prioritize formation speed when needed and traffic efficiency at other times.

  • Dynamic Lane Management: Instead of fixed UGV-dedicated lanes, this module adaptively opens or closes these lanes based on real-time traffic flow and speed. This ensures that dedicated lanes are used efficiently, especially in high-flow scenarios, and avoids underutilization when traffic is light.
  • Hybrid Microscopic-Macroscopic Traffic Model: The system integrates detailed vehicle dynamics (microscopic) with a cell transmission model (macroscopic) to capture how individual UGV actions affect overall traffic flow and vice-versa.

Datasets

Not discussed in the paper. The authors rely on simulations to evaluate their proposed framework.

Key Insights and Ideas

  • Adaptive weights are crucial: The paper highlights that a fixed weighting scheme for MPC objectives is suboptimal. Dynamically adjusting weights based on traffic conditions and formation phase significantly improves both platoon formation speed and overall traffic efficiency.
  • Dynamic lane management is key to efficiency: Simply dedicating lanes to UGVs isn't always optimal. Adaptively opening and closing these lanes based on traffic demand prevents underutilization and integrates platoons more smoothly into the broader traffic flow.
  • Decomposition simplifies complexity: Breaking the complex platoon control problem into two distinct MPC stages (pursuit and formation) makes the optimization problem more manageable and allows for tailored strategies for each phase.
  • Platooning benefits increase with UGV penetration: Higher percentages of UGVs in the traffic stream lead to better platoon formation and overall traffic performance, as UGVs respond more predictably to control actions.

Key Results

  • Platoon Formation Time: The adaptive weighting strategy reduced platoon formation time by 41.6% compared to fixed-weight strategies.
  • Traffic Speed Impact: This speedup came with a minimal reduction in average traffic speed of only 1.29%.
  • Robustness: The framework demonstrated robustness against sensing noise (up to 15%), V2X latency (up to 150 ms), and unexpected HDV cut-ins, maintaining safety (minimum Time-to-Collision above 2.0 s).
  • Dynamic Lane Management: This strategy enabled longer and more stable platoons, particularly in high-flow environments, without significant efficiency losses.
  • Scalability: Simulations showed the approach remains effective for platoons of three and five vehicles on curved roads.

Limitations and Future Work

  • Simulation-based: The results are based on simulations, and real-world validation is needed to confirm performance under actual sensor, communication, and human interaction uncertainties.
  • Heterogeneous Systems: The paper suggests integrating other unmanned systems (like UAVs) for enhanced coordination.
  • Learning-based Controllers: Future work could explore learning-based adaptive controllers for even greater scalability and adaptability to rapidly changing environments.
  • Non-fully Connected Environments: Developing trajectory planning models for scenarios where vehicles are not fully connected is another avenue for future research.

Summary updated

Abstract

Cooperative longitudinal–lateral trajectory optimization is essential for unmanned ground vehicle (UGV) platoons to improve safety, capacity, and efficiency. However, existing approaches often face unstable formation under low penetration rates and rely on fragmented control strategies. This study develops a cooperative longitudinal–lateral trajectory tracking framework tailored for UGV platooning, embedded in a hierarchical control architecture. Dual-stage multi-objective Model Predictive Control (MPC) is proposed, decomposing trajectory planning into pursuit and platooning phases. Each stage employs adaptive weighting to balance platoon efficiency and traffic performance across varying operating conditions. Furthermore, a traffic-aware organizational module is designed to enable the dynamic opening of UGV-dedicated lanes, ensuring that platoon formation remains compatible with overall traffic flow. Simulation results demonstrate that the adaptive weighting strategy reduces the platoon formation time by 41.6% with only a 1.29% reduction in the average traffic speed. In addition, the dynamic lane management mechanism yields longer and more stable UGV platoons under different penetration levels, particularly in high-flow environments. The proposed cooperative framework provides a scalable solution for advancing UGV platoon control and demonstrates the potential of unmanned systems in future intelligent transportation applications.
BibTeX
@inproceedings{wang2025cooperative,
  author = {Shunchao Wang and Zhigang Wu and Yonghui Su},
  title = {{A Cooperative Longitudinal-Lateral Platoon Control Framework with Dynamic Lane Management for Unmanned Ground Vehicles Based on a Dual-Stage Multi-Objective MPC Approach}},
  booktitle = {Drones},
  year = {2025},
  doi = {10.3390/drones9100711},
}
Figure