Online Identification of Individual Driver Steering Behaviour and Experimental Results

Abstract

Using switched systems, we model individual driver steering behaviour from a new point of view. This approach allows to incorporate the idea of human motion being built up by an individual and limited repertoire of learned patterns. The identification of the generating subsystem parameters of such individual motion primitives solely on measured output data requires a new identification method. We propose an algorithm using a multi-step model output error criterion and discuss different implementations in detail. We show that this method is capable of tracking real measurement data of driver steering motion trajectories with low model orders and number of switches respectively. The presented method is online-capable. Experimental driving results proof the concept.Index Terms-individual driver steering model, motion primitives, switched systems, identification algorithm.

Gray-Box Driver Modeling and Prediction: Benefits of Steering Primitives

“…according to [13]. These generate T samples of an output trajectory {w k } T k=1 , w k ∈ R whereas λ k ∈ M, M = {1 .…”

“…In [13], [14] a multi-step prediction error-based algorithm for the identification of movemes from the measured output data has been presented and validated experimentally, even for noisy measurements. Thereby the order n a (λ) is assumed to be uniform for all subsystems.…”

“…In this specific scope as a gray-box model for the neuromuscular system of the driver, we model these primitives as autonomous dynamic systems and call them movemes in line with [12]. A significant benefit of the movemes is that they can comparatively easy be identified using the algorithms submitted in [13] [14] and yield guarantees regarding correct identification. In previous work [15] [16] we proposed a driver model for the steering task which utilizes these movemes to generate a trajectory of driver steering motion.…”

Individual Driver Modeling via Optimal Selection of Steering Primitives

“…We assume that the driver specific parameter set A i and b i , ∀i ∈ I of the movemes are identified from real steering wheel trajectories and are known. These identification tasks can be done using Diehm et al [2013b]. In this paper we are then confronted with the following question: For a given maneuver, which moveme of the individual set has to be chosen such that the car performs the maneuver.…”