A Probabilistic Framework for Imitating Human Race Driver Behavior

Löckel, Stefan; Peters, Jan; Vliet, Peter Van de

doi:10.1109/lra.2020.2970620

Cited by 13 publications

(7 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The measures were used to objectively detect differences between drivers. Löckel et al [9] proposed a probabilistic framework for imitating race drivers. They defined a set of metrics based on driver inputs, for example steering or braking aggressiveness to evaluate their framework.…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Race Driver Evaluation at a Driving Simulator using a physical Model and a Machine Learning Approach

Schleinitz¹,

Schwarzhuber²,

Wörle³

et al. 2022

Preprint

View full text Add to dashboard Cite

Professional race drivers are still superior to automated systems at controlling a vehicle at its dynamic limit. Gaining insight into race drivers' vehicle handling process might lead to further development in the areas of automated driving systems. We present a method to study and evaluate race drivers on a driver-in-the-loop simulator by analysing tire grip potential exploitation. Given initial data from a simulator run, two optimiser based on physical models maximise the horizontal vehicle acceleration or the tire forces, respectively. An overall performance score, a vehicle-trajectory score and a handling score are introduced to evaluate drivers. Our method is thereby completely track independent and can be used from one single corner up to a large data set. We apply the proposed method to a motorsport data set containing over 1200 laps from seven professional race drivers and two amateur drivers whose lap times are 10 − 20% slower. The difference to the professional drivers comes mainly from their inferior handling skills and not their choice of driving line. A downside of the presented method for certain applications is an extensive computation time. Therefore, we propose a Long-short-term memory (LSTM) neural network to estimate the driver evaluation scores. We show that the neural network is accurate and robust with a root-mean-square error between 2 − 5% and can replace the optimisation based method. The time for processing the data set considered in this work is reduced from 68 hours to 12 seconds, making the neural network suitable for real-time application.

show abstract

Section: Related Workmentioning

confidence: 99%

“…L( ẋ), c( ẋ) and ceq( ẋ) can be nonlinear functions [23]. For the problem specified in this work, ceq( ẋ) contains the vehicle model's momentum equalities (3,4,5) as well as equality constraints (7,8,9,10). The inequalities contained in c( ẋ) are described by equations (11,12,13).…”

Section: Optcog Optimisationmentioning

confidence: 99%

Race Driver Evaluation at a Driving Simulator using a physical Model and a Machine Learning Approach

Schleinitz¹,

Schwarzhuber²,

Wörle³

et al. 2022

Preprint

View full text Add to dashboard Cite

show abstract

“…The Probabilistic Modeling of Driver Behavior (ProMoD) framework was demonstrated to be capable of completing full laps with a competitive performance by mimicking professional race drivers in a realistic HDiL simulation environment [2], [3]. The data-based and modular approach learns distributions of driving lines represented by Probabilistic Movement Primitives (ProMPs) [28], [29] and trains a recurrent neural network on human race driver data in a supervised fashion.…”

Section: B Related Workmentioning

confidence: 99%

“…It consists of multiple modules as visualized in Figure 3, where each of these modules represents fundamental steps in the decision-making process of a human race driver. [2], [3] In the following, the existing architecture is shortly summarized in order to provide a proper foundation for the subsequent development of the novel generalization and adaptation methods:…”

Section: B Promodmentioning

confidence: 99%

“…2) vehicle dynamics are highly nonlinear and race cars are usually driven at the handling limits, and thus difficult to control, 3) each driver exhibits an individual driving style, 4) the generalization and adaptation mechanisms are complex and difficult to incorporate in a driver model. While challenges 1-3 have been successfully addressed in recent research [2], [3], the problem of understanding and imitating the human adaptation process remains unsolved. With our work, we intend to identify and understand the most important adaptation and learning techniques mastered by professional race drivers, contribute to the modeling of driver behavior by developing two methods to mimic this behavior, and evaluate the proposed methodology within a realistic Human-Driver-in-the-Loop (HDiL) simulator as shown in Figure 1.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

An Adaptive Human Driver Model for Realistic Race Car Simulations

Löckel¹,

Ju²,

Schaller³

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

Engineering a high-performance race car requires a direct consideration of the human driver using real-world tests or Human-Driver-in-the-Loop simulations. Apart from that, offline simulations with human-like race driver models could make this vehicle development process more effective and efficient but are hard to obtain due to various challenges. With this work, we intend to provide a better understanding of race driver behavior and introduce an adaptive human race driver model based on imitation learning. Using existing findings and an interview with a professional race engineer, we identify fundamental adaptation mechanisms and how drivers learn to optimize lap time on a new track. Subsequently, we use these insights to develop generalization and adaptation techniques for a recently presented probabilistic driver modeling approach and evaluate it using data from professional race drivers and a state-of-the-art race car simulator. We show that our framework can create realistic driving line distributions on unseen race tracks with almost human-like performance. Moreover, our driver model optimizes its driving lap by lap, correcting driving errors from previous laps while achieving faster lap times. This work contributes to a better understanding and modeling of the human driver, aiming to expedite simulation methods in the modern vehicle development process and potentially supporting automated driving and racing technologies.

show abstract