SafeVRU

From VRA
Jump to: navigation, search


SafeVRU
Safe Interaction of Automated Vehicles with Vulnerable Road Users
Random logo.JPG
General information
Category
The Netherlands STW
From 2016 to 2020 Ongoing
Contact
[1]
Project Coordinator

TUD

SafeVRU will develop sensing, intent recognition, and vehicle control strategies for the interaction of automated vehicles with vulnerable road users (VRU) such as pedestrians and cyclists.

Keypoints

The project will enhance VRU detection and classification by fusing vision and radar. The benefit of cooperative localization will be examined for added robustness; especially in case of visibility obstructions. Novel VRU intent recognition and path prediction methods will be developed using kinematics, gestures, visual focus and scene context. Taking into account the complex interaction with multiple VRU, real time adaptive path planning strategies will be developed to minimize risk, with vehicle motion being intuitive and predictable for VRU.

Outcomes

The SafeVRU technologies will be validated using real world data, virtual reality labs and fully automated vehicles (WEpods) in the province of Gelderland.

Related Projects

DAVI, WEpods

Partners

  • TU Delft
  • University of Amsterdam
  • TNO
  • SWOV
  • NXP
  • 2getthere
  • RDW
  • Province of Gelderland

Budget

Funded under the Technology Foundation STW with a budget of: €688.057

Keywords