Advertisement

ATZ worldwide

, Volume 121, Issue 4, pp 26–31 | Cite as

Pedestrian Intention and Gesture Classification Using Neural Networks

  • Thomas Brohm
  • Karl Haupt
  • Robert Thiel
Cover Story
  • 15 Downloads

A comprehensive understanding of the situation surrounding a vehicle - the so-called environmental model - is a prerequisite for safe automated driving. In the interaction between pedestrians and automated vehicles, pedestrian detection plays an important role. With the help of artificial intelligence and machine learning, Continental is working on the detection of pedestrians and their intention to move.

Algorithmic Paradigm Shift

Whether a vehicle is controlled by the driver, supports the driver in assisted driving or drives automatically, the processes are always the same: The driver or vehicle has to sense the environment and the current vehicle status (sense), plan actions (plan) and act using the various actuators and control systems in the vehicle (act). This Sense-Plan-Act chain of effects is an important element on the road toward Vision Zero - the vision of accident-free driving. What constitutes a simple task for a human driver can be a real challenge for a camera...

Copyright information

© Springer Fachmedien Wiesbaden GmbH, part of Springer Nature 2019

Authors and Affiliations

  • Thomas Brohm
    • 1
  • Karl Haupt
    • 1
  • Robert Thiel
    • 1
  1. 1.ContinentalLindauGermany

Personalised recommendations