Methodology for designing collision avoidance systems based on multiobjective criteria

Nobukawa, K. and Gordon, Timothy J. (2014) Methodology for designing collision avoidance systems based on multiobjective criteria. In: 21st World Congress on Intelligent Transport Systems: Reinventing Transportation in Our Connected World, ITSWC 2014, 7-11 September 2014, Detroit, Michigan, USA.

Full text not available from this repository.

Item Type:Conference or Workshop contribution (Paper)
Item Status:Live Archive

Abstract

This paper proposes and demonstrates a methodology for optimizing the performance of collision avoidance systems. The case considered is for intersection left turns, and the method used is based on multiobjective optimization through computer simulation. More specifically, Pareto optimality criteria are applied to Monte-Carlo simulations using a vehicle/driver integrated dynamic model. The following steps are involved in the methodology: 1) conflict analyses of actual left turn events for human drivers' gap acceptance behavior, 2) development of the simulation model, 3) preparation of virtual left turn events by Monte-Carlo simulations using randomized values for the driver parameters and vehicle initial conditions, 4) evaluation of the system performance by conducting conflict analyses of the virtual left turns for various values for the design variables, 5) construction of the Pareto optimal solutions and corresponding objective sets based on the previous step, and 6) selection of an appropriate operating point from within the objective set. This methodology is general and can be applied to other conflict scenarios.

Additional Information:Conference Code:111602
Keywords:Airships, Collision avoidance, Intelligent systems, Intersections, Monte Carlo methods, Pareto principle, Research aircraft, Safety devices, Traffic control, Virtual reality, Collision avoidance systems, Conflict analysis, Driver modeling, Initial conditions, Left turns, Pareto optimal solutions, Pareto-optimality, Simulation model, Multiobjective optimization
Subjects:H Engineering > H330 Automotive Engineering
H Engineering > H331 Road Vehicle Engineering
Divisions:College of Science > School of Engineering
ID Code:17532
Deposited On:27 May 2015 14:45

Repository Staff Only: item control page