Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Shared Control for Vehicle Lane-Changing with Uncertain Driver Behaviors

Jiamin Wu, Chenguang Zhao, Huan Yu

TL;DR

This work tackles lane-change stability under uncertain driver behavior by modeling driver longitudinal actions as a continuous-time hidden Markov process and designing a human–automation shared controller. A nominal stabilizing controller guarantees stochastic $L_2$ string stability via LMIs, even with imperfect mode estimation, and a Minimal Intervention Controller (MIC) adds an automated-effort penalty to balance stability with driver authority. Simulations on NGSIM and TGSIM show the nominal controller reduces disturbances and speeds up lane changes, while MIC reduces automation effort and improves comfort with a tunable trade-off governed by $eta$. The results highlight how shared control can enhance stability, efficiency, and driver acceptance in uncertain traffic, with practical implications for SAE Level 2 and higher deployments.

Abstract

Lane changes are common yet challenging driving maneuvers that require continuous decision-making and dynamic interaction with surrounding vehicles. Relying solely on human drivers for lane-changing can lead to traffic disturbances due to the stochastic nature of human behavior and its variability under different task demands. Such uncertainties may significantly degrade traffic string stability, which is critical for suppressing disturbance propagation and ensuring smooth merging of the lane-changing vehicles. This paper presents a human-automation shared lane-changing control framework that preserves driver authority while allowing automated assistance to achieve stable maneuvers in the presence of driver's behavioral uncertainty. Human driving behavior is modeled as a Markov jump process with transitions driven by task difficulty, providing a tractable representation of stochastic state switching. Based on this model, we first design a nominal stabilizing controller that guarantees stochastic ${L}_2$ string stability under imperfect mode estimation. To further balance performance and automated effort, we then develop a Minimal Intervention Controller (MIC) that retains acceptable stability while limiting automation. Simulations using lane-changing data from the NGSIM dataset verify that the nominal controller reduces speed perturbations and shorten lane-changing time, while the MIC further reduces automated effort and enhances comfort but with moderate stability and efficiency loss. Validations on the TGSIM dataset with SAE Level 2 vehicles show that the MIC enables earlier lane changes than Level 2 control while preserving driver authority with a slight stability compromise. These findings highlight the potential of shared control strategies to balance stability, efficiency, and driver acceptance.

Shared Control for Vehicle Lane-Changing with Uncertain Driver Behaviors

TL;DR

This work tackles lane-change stability under uncertain driver behavior by modeling driver longitudinal actions as a continuous-time hidden Markov process and designing a human–automation shared controller. A nominal stabilizing controller guarantees stochastic string stability via LMIs, even with imperfect mode estimation, and a Minimal Intervention Controller (MIC) adds an automated-effort penalty to balance stability with driver authority. Simulations on NGSIM and TGSIM show the nominal controller reduces disturbances and speeds up lane changes, while MIC reduces automation effort and improves comfort with a tunable trade-off governed by . The results highlight how shared control can enhance stability, efficiency, and driver acceptance in uncertain traffic, with practical implications for SAE Level 2 and higher deployments.

Abstract

Lane changes are common yet challenging driving maneuvers that require continuous decision-making and dynamic interaction with surrounding vehicles. Relying solely on human drivers for lane-changing can lead to traffic disturbances due to the stochastic nature of human behavior and its variability under different task demands. Such uncertainties may significantly degrade traffic string stability, which is critical for suppressing disturbance propagation and ensuring smooth merging of the lane-changing vehicles. This paper presents a human-automation shared lane-changing control framework that preserves driver authority while allowing automated assistance to achieve stable maneuvers in the presence of driver's behavioral uncertainty. Human driving behavior is modeled as a Markov jump process with transitions driven by task difficulty, providing a tractable representation of stochastic state switching. Based on this model, we first design a nominal stabilizing controller that guarantees stochastic string stability under imperfect mode estimation. To further balance performance and automated effort, we then develop a Minimal Intervention Controller (MIC) that retains acceptable stability while limiting automation. Simulations using lane-changing data from the NGSIM dataset verify that the nominal controller reduces speed perturbations and shorten lane-changing time, while the MIC further reduces automated effort and enhances comfort but with moderate stability and efficiency loss. Validations on the TGSIM dataset with SAE Level 2 vehicles show that the MIC enables earlier lane changes than Level 2 control while preserving driver authority with a slight stability compromise. These findings highlight the potential of shared control strategies to balance stability, efficiency, and driver acceptance.

Paper Structure

This paper contains 25 sections, 4 theorems, 65 equations, 10 figures, 3 tables.

Table of Contents

  1. Introduction
  2. Lane-changing control on automated vehicles
  3. Shared control methods and applications
  4. Contributions
  5. Problem formulation
  6. Vehicle motion modeling and system descriptions
  7. Human behavior modeling
  8. Stabilization of assistive controller
  9. Linearization
  10. Stochastic $\mathcal{L}_2$ string stability
  11. Minimal Intervention Controller Design
  12. Balance between stability performance and automated effort
  13. Controller synthesis with minimal intervention
  14. Numerical results
  15. Simulation setup
  16. ...and 10 more sections

Key Result

Lemma 1

Li2019stringstability For the lane-changing system eq:system2, when the initial condition is $\tilde{x}(0)=0$, it is stochastically $\mathcal{L}_2$ string stable, if and only if there exists $\gamma\le 1$ such that:

Figures (10)

  • Figure 1: The ego-vehicle adjusts its speed and gap, then changes to the target lane when it has a suitable gap with both the leader vehicle and the follower vehicle. The entire figure illustrates the proposed shared control framework: the human driver generates inputs based on observations, while the assistive controller provides assistance through either the nominal stabilizing controller or the minimal intervention design, both of which adapt to the estimated human mode $\hat{\eta}$. The combined input $u=u_{{\scaleto{\mathrm{H}}{3.5pt}}}+u_{{\scaleto{\mathrm{AV}}{3.5pt}}}$ drives the ego-vehicle, ensuring disturbance attenuation and maintaining driver authority under stochastic behavior transitions.
  • Figure 2: Mode evolution under the two schemes. The left panel shows the TD-based true mode $\eta(t)$ when no automated assistance is applied. The right panel plots the true mode and the observed mode produced by the estimator under nominal shared control. With AV assistance, the high TD duration is shortened, which implies lower driving difficulties for human drivers.
  • Figure 3: Trajectories of vehicles under human-only control and nominal shared control.
  • Figure 4: Trajectories of vehicles under shared control with MIC.
  • Figure 5: Comparison of input difference $\Delta u = u_{{\scaleto{\mathrm{H}}{3.5pt}}}-u_{{\scaleto{\mathrm{AV}}{3.5pt}}}$ and control inputs $u = u_{{\scaleto{\mathrm{H}}{3.5pt}}}+u_{{\scaleto{\mathrm{AV}}{3.5pt}}}$ between nominal and minimal intervention controllers.
  • ...and 5 more figures

Theorems & Definitions (10)

  • Definition 1: Stochastic $\mathcal{L}_2$ string stability Li2019stringstability
  • Lemma 1
  • Theorem 1
  • proof
  • Remark 1
  • Lemma 2: Stochastic $\mathcal{L}_2$ performance under augmented output
  • Theorem 2: Stochastic $\mathcal{L}_2$ string stability with minimal intervention
  • proof
  • Remark 2
  • Remark 3