Generalized two-point visual control model of human steering for accurate state estimation

TL;DR

This work addresses the challenge of estimating vehicle state under lane-center ambiguity by merging human steering input with autonomous sensors through a generalized two-point visual control model. The generalized model, which extends the classic two-point framework with higher-order autoregression and a lateral-momentum term, achieves substantial improvements in predicting human steering and enables accurate lateral-state estimation when lane markings are biased or unclear. By coupling this model with a Gaussian-mixture extended Kalman filter in a human-as-advisor framework, the authors demonstrate robust state estimation across multiple drivers and track geometries, achieving average lateral error below 0.25 m and maximum RMSE around 0.25 m. The approach has practical implications for safer human-autonomy collaboration in steering tasks and for improving lane-centering performance under perceptual ambiguity.

Abstract

We derive and validate a generalization of the two-point visual control model, an accepted cognitive science model for human steering behavior. The generalized model is needed as current steering models are either insufficiently accurate or too complex for online state estimation. We demonstrate that the generalized model replicates specific human steering behavior with high precision (85\% reduction in modeling error) and integrate this model into a human-as-advisor framework where human steering inputs are used for state estimation. As a benchmark study, we use this framework to decipher ambiguous lane markings represented by biased lateral position measurements. We demonstrate that, with the generalized model, the state estimator can accurately estimate the true vehicle state, providing lateral state estimates with under 0.25 m error on average across participants. However, without the generalized model, the estimator cannot accurately estimate the vehicle's lateral state.

Figures (9)

• Figure 1: State estimation with human-as-advisor.
• Figure 2: Ambiguous lane center, adapted from Shaw2018.
• Figure 3: Near-point and far-point illustration with kinematic bicycle model
• Figure 4: Steering performance comparison on validation data.
• Figure 5: Autocorrelation of $\delta_h-\hat{\delta}_h$ for the generalized model.