Aircraft Collision Avoidance Systems: Technological Challenges and Solutions on the Path to Regulatory Acceptance

TL;DR

The paper surveys the development of aircraft collision avoidance systems, focusing on surveillance, decision-making, validation, and regulatory acceptance, with emphasis on ACAS X and TCAS. It presents ACAS X as an $MDP$-based, cost-structured decision framework that uses belief states and $QMDP$-style lookup to handle outcome and state uncertainties, respectively. Validation is anchored in encounter modeling, Monte Carlo and importance sampling for rare-event estimation, stress testing, formal methods, and iterative flight testing, all tied to safety metrics like $NMAC$ and risk ratios. Regulatory acceptance hinges on harmonized standards (RTCA/EUROCAE MOPS, DO-185B/DO-385A), certification pathways, and societal acceptance through training and monitoring, with future work extending coverage to rotorcraft, general aviation, and integration with higher-level autonomy.

Abstract

Aircraft collision avoidance systems is critical to modern aviation. These systems are designed to predict potential collisions between aircraft and recommend appropriate avoidance actions. Creating effective collision avoidance systems requires solutions to a variety of technical challenges related to surveillance, decision making, and validation. These challenges have sparked significant research and development efforts over the past several decades that have resulted in a variety of proposed solutions. This article provides an overview of these challenges and solutions with an emphasis on those that have been put through a rigorous validation process and accepted by regulatory bodies. The challenges posed by the collision avoidance problem are often present in other domains, and aircraft collision avoidance systems can serve as case studies that provide valuable insights for a wide range of safety-critical systems.

Figures (9)

• Figure 1: Worldwide jet transport flight hours and mid-air collision rates from 1970--2025. Mid-air collision events are indicated with red markers derived from data provided by BoeingAccidentSummary2025BoeingAccidentSummary2025.
• Figure 2: Simplified TCAS logic for an example scenario with a minimum required separation of 400. In the first step, the logic selects between a climb or descend sense by projecting the results of two maneuver templates assuming a five second pilot response delay and selecting the sense of the maneuver (in this case, down) that results in the greatest vertical separation from the intruder. In the second step, the logic projects the results of four different vertical RAs (limit climb to 500min, do not climb, descend at 1500min, and descend at 2500min). It then selects the minimum-strength RA that results in at least 400 of vertical separation from the intruder. In this case, the logic will advise the pilot to descend at 1500min (shown in solid green).
• Figure 3: Two types of uncertainty address by ACAS X. The left side shows the outcome uncertainty, which is addressed during the optimization of the logic table. The right side shows the state uncertainty, which is addressed during the lookup process.
• Figure 4: State variables used in the ACAS X MDP. The vertical logic state consists of the relative altitude $h$, ownship vertical rate $\dot h_0$, intruder vertical rate $\dot h_1$, the previous advisory (not shown), and the time to loss of horizontal separation $\tau$. The horizontal logic state consists of the range $r$, bearing $\theta$, relative heading $\psi$, ownship horizontal speed $v_0$, intruder horizontal speed $v_1$, the previous advisory (not shown), and the time to loss of vertical separation $\tau$.
• Figure 5: Slices of the optimal policy for a notional example with three possible actions. In the plot on the left, both aircraft are in level flight, while in the plot on the right, the ownship is climbing. The previous action is fixed at no advisory in both plots. The colors represent the optimal action at each state, and the red aircraft represents the location of the intruder aircraft.