Virtual Reality Setup. You can the head up display and the simulated naturalistic environment. (click on image to see an animation)
Can a self-driving vehicle be moral, act like humans do, or act like humans expect humans to?
To answer this we perform experiments in virtual reality to allow to study human behavior in simulated road traffic scenarios and use machine learning to model human bevaiour.
The participants were asked to drive a car in a typical suburban neighborhood on a foggy day when they experienced unexpected unavoidable dilemma situations with inanimate objects, animals, and humans and had to decide which was to be spared. The results were conceptualized by statistical models leading to rules, with an associated degree of explanatory power to explain the observed behavior. The research showed that moral decisions in the conﬁned scope of unavoidable traffic collisions can be explained well, and modeled, by a single value-of-life for every human, animal, or inanimate object.
L.R. Sütfeld, R. Gast, P. König, G. Pipa, 'Using Virtual Reality to Assess Ethical Decisions in Road Traffic Scenarios: Applicability of Value-of-Life-Based Models and Influences of Time Pressure', Front. Behav. Neurosci., 05 July 2017 https://doi.org/10.3389/fnbeh.2017.00122
A. Skulmowski, A. Bunge, K. Kaspar, G. Pipa, 'Forced-choice decision-making in modified trolley dilemma situations: a virtual reality and eye tracking study', Front. Behav. Neurosci., 16 December 2014 | https://doi.org/10.3389/fnbeh.2014.00426
MA Wächter, A Faulhaber, F Blind, S Timm, A Dittmer, LR Sütfeld, A Stephan, G Pipa, P König, 'Human decisions in moral dilemmas are largely described by Utilitarianism: virtual car driving study provides guidelines for ADVs', arXiv preprint arXiv:1706.07332