Event Abstract

Neural Signatures Enhance Emergency Braking Intention Detection during Simulated Driving

  • 1 Berlin Institute of Technology, Machine Learning Group, Germany
  • 2 Bernstein Focus Neurotechnology, BFNT- B, Germany
  • 3 Fraunhofer FIRST, IDA, Germany
  • 4 Humboldt University and the Free University, Charité University Medicine, Germany

In many safety-critical areas such as individual car traffic, human performance has become the limiting factor. Modern driving assistance systems can detect potential upcoming crashes and take preparatory measures or even initiate automatic emergency braking.We conducted a driving simulator study (N=20) using a customized version of the open-source racing software TORCS. The experiment comprised 3 blocks (45 minutes each) of driving, in which the subjects had to tightly follow a computer-controlled car at a speed of 100 km/h. While subjects were within the desired maximal distance of 20 m, the preceding car occasionally (20-40 seconds ISI, randomized) performed sudden brakings, forcing the subject to immediately brake as well in order to avoid a crash.Various ERP components predictive of emergency braking occasions could be observed between the initial flashing of the preceding car's brake light and the onset of braking: a VEP related to low-level visual processing of the brake light stimulus and a P3 reflecting the rareness and importance of this event. Further, a readiness potential (LRP) was present, building up before the actual movement. While radar-derived features (distance and acceleration) were the first to indicate a critical situation at all, neural parameters were the earliest to reflect driver's intent to brake, followed by electromyographic, gas pedal and braking pedal derived features.All the observed ERPs occur frequently during normal driving. However, our analysis shows that their co-occurrence robustly characterizes certain critical situations. Furthermore, inclusion of (neuro-) physiological features lead to an increase of detection performance even in early stages of critical situations, showing that such features carry information about the driver's intent that is strictly additive to behavioural and technical parameters.

Figure 1:Grand-average ERPs relative to braking onset

Figure 1

Keywords: computational neuroscience

Conference: Bernstein Conference on Computational Neuroscience, Berlin, Germany, 27 Sep - 1 Oct, 2010.

Presentation Type: Presentation

Topic: Bernstein Conference on Computational Neuroscience

Citation: Haufe S, Treder M, Gugler M, Sagebaum M, Ewald A, Curio G and Blankertz B (2010). Neural Signatures Enhance Emergency Braking Intention Detection during Simulated Driving. Front. Comput. Neurosci. Conference Abstract: Bernstein Conference on Computational Neuroscience. doi: 10.3389/conf.fncom.2010.51.00026

Received: 20 Sep 2010; Published Online: 23 Sep 2010.

* Correspondence: Dr. Stefan Haufe, Berlin Institute of Technology, Machine Learning Group, Berlin, Germany, stefan.haufe@tu-berlin.de

© 2007 - 2017 Frontiers Media S.A. All Rights Reserved