NASA Logo

NTRS

NTRS - NASA Technical Reports Server

Collections About News Help Login
Back to Results
Adaptive Stress Testing of Airborne Collision Avoidance SystemsThis 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.
Document ID
20160005033
Acquisition Source
Ames Research Center
Document Type
Conference Paper
Authors
Lee, Ritchie (SGT, Inc. Moffett Field, CA, United States)
Kochenderfer, Mykel J. (Stanford Univ. Stanford, CA, United States)
Mengshoel, Ole J. (Carnegie-Mellon Univ. Moffett Field, CA, United States)
Brat, Guillaume P. (Carnegie-Mellon Univ. Moffett Field, CA, United States)
Owen, Michael P. (Massachusetts Inst. of Tech. Lexington, MA, United States)
Date Acquired
April 14, 2016
Publication Date
September 13, 2015
Subject Category
Aircraft Design, Testing And Performance
Report/Patent Number
ARC-E-DAA-TN26126
Meeting Information
Meeting: Digital Avionics Systems Conference
Location: Prague
Country: Czechoslovakia
Start Date: September 13, 2015
End Date: September 17, 2015
Sponsors: Institute of Electrical and Electronics Engineers
Funding Number(s)
CONTRACT_GRANT: NNA14AA60C
Distribution Limits
Public
Copyright
Public Use Permitted.
Keywords
Verification and Validation
Reinforcement Learning
ACAS X

Related Records

There are no records associated with this record.
No Preview Available