Autonomous Driving in Mixed-Traffic Urban Environments

Abstract

This study focuses on a number of control, sensing and coordination aspects of autonomous navigation in real-life urban traffic. On the control side, by expanding the inherent hierarchy of the hybrid-state system formulation, a highly structured yet modular architecture was developed to connect various traffic elements. The feasibility of coordination under vehicle-to-vehicle and vehicle-to-infrastructure communication was investigated through experiments with multiple vehicles and roadside units. Further experimentation through a 100-vehicle collaborative demo is also under development. The necessary levels of situational-awareness abstractions for different tiers of the proposed hierarchy were studied on stochastic hybrid-state system estimators. Potential distributed control schemes were investigated through overlapping decompositions and hybrid-state system controllers were analyzed for safety through software verification and model checking.

On the sensing side, the researchers investigated sensory data processing, filtering and sensor fusion methods for autonomous vehicles operating in real-life, urban environments with human and machine drivers, and pedestrians. Extended Kalman Filters were used to develop decentralized data fusion algorithms for communicating vehicles; Particle Filters were improved by assigning trust/confidence values in order to overcome faulty/compromised sensors; and the computational cost of particle filters were distributed by parallelizing the load using the developed Shared-Memory Systematic Resampling algorithm.

Award ID: 0931669

  • 0931669
  • Automotive
  • CPS Domains
  • Transportation Systems Sector
  • Control
  • Modeling
  • Systems Engineering
  • Critical Infrastructure
  • Wireless Sensing and Actuation
  • Robotics
  • Transportation
  • CPS Technologies
  • Foundations
  • National CPS PI Meeting 2012
  • 2012
  • Poster
  • Academia
  • CPS PI MTG 12 Posters & Abstracts
Submitted by Umit Ozguner on Thu, 10/04/2012 - 11:56