Adaptive stress testing of airborne collision avoidance systems
- 1 September 2015
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- No. 21557195,p. 6C2-1
Abstract
This paper presents a scalable method to efficiently search for the most likely state trajectory leading to an event given only a simulator of a system. Our approach uses a reinforcement learning formulation and solves it using Monte Carlo Tree Search (MCTS). The approach places very few requirements on the underlying system, requiring only that the simulator provide some basic controls, the ability to evaluate certain conditions, and a mechanism to control the stochasticity in the system. Access to the system state is not required, allowing the method to support systems with hidden state. The method is applied to stress test a prototype aircraft collision avoidance system to identify trajectories that are likely to lead to near mid-air collisions. We present results for both single and multi-threat encounters and discuss their relevance. Compared with direct Monte Carlo search, this MCTS method performs significantly better both in finding events and in maximizing their likelihood.Keywords
This publication has 6 references indexed in Scilit:
- Optimizing the Next Generation Collision Avoidance System for Safe, Suitable, and Acceptable Operational PerformanceAir Traffic Control Quarterly, 2013
- A Survey of Monte Carlo Tree Search MethodsIEEE Transactions on Computational Intelligence and AI in Games, 2012
- Continuous Upper Confidence Trees2011
- A decision-theoretic approach to developing robust collision avoidance logic2010
- Airspace Encounter Models for Estimating Collision RiskJournal of Guidance, Control, and Dynamics, 2010
- Reinforcement Learning: An IntroductionIEEE Transactions on Neural Networks, 1998