Next-Generation Airborne Collision Avoidance System

By Mykel J. Kochenderfer, Jessica E. Holland, James P. Chryssanthacopoulos. 2012

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Citations in the corpus

Efficient Formal Safety Analysis of Neural Networks (2018, background): “Unmanned aircraft alert systems (ACAS Xu) [15] are networks advising decisions for aircraft, which is on schedule to be installed in over 30,000 passengers and cargo aircraft worldwide [20] and US Navy’s fleets [1].”
Formal Security Analysis of Neural Networks using Symbolic Intervals (2018, methodology): “We focus our evaluation on the latest system in this category called ACAS Xu for unmanned system [23], as Reluplex also evaluated on the same system where ours substantially outperform all the models tested in Reluplex.”
Reluplex: An Efficient SMT Solver for Verifying Deep Neural Networks (2017, methodology): “This system adopts an approach that involves solving a partially observable Markov decision process to optimize the alerting logic and further reduce the probability of midair collisions, while minimizing unnecessary alerts [18,19,21].”
Learning Plannable Representations with Causal InfoGAN (2018, background): “Indeed, most planning successes to date (e.g., [35, 25, 47]) relied on a human-designed state representation, and manually designed the perception of the state from the observation.”
Differential Adaptive Stress Testing of Airborne Collision Avoidance Systems (2018, unknown): “While studies have shown that NMACs are extremely rare, the risk of NMAC cannot be completely eliminated due to factors such as surveillance noise, pilot response delay, and the need for an acceptable alert rate [2].”
A Formally Verified Hybrid System for the Next-Generation Airborne Collision Avoidance System (2015, methodology): “ACAS X is designed to avoid Near Mid-Air Collisions (NMACs), situations where two aircraft come within rp = 500 ft horizontally and hp = 100 ft vertically [13] of each other.”
Developing Decision Aids to Enable Human Spaceflight Autonomy (2017, background): “Examples of developments in this area include collision detection and avoidance (Kochenderfer, Holland, and Chryssanthacopoulos 2012), flight path situational awareness in modern glass cockpit aircraft (Neville and Dey 2012), emergency landing planning (Meuleau et al. 2011), and autonomous loss of…”
The Adaptive Stress Testing Formulation (2020, background): “1) Problem: The next-generation Airborne Collision Avoidance System (ACASX) [3] gives instructions to pilots when multiple planes are approaching each other.”
Policy compression for aircraft collision avoidance systems (2016, background): “Recent work on formulating the problem of collision avoidance as a partially observable Markov decision process (POMDP) has led to the development of the ACAS X family of collision avoidance systems [2], [3], [4].”