Results show that training significantly reduced error rates and response times, in line with theories of expertise acquisition, and support the relevance of cognitive training for preparing aircrews for future collaborative combat involving ACPs.
Abstract
High-tempo domains such as defense, medicine, and transportation have been transformed by AI-based systems that accelerate information processing and decision support. Modern air combat reflects this evolution, with automated systems continuously generating tactical options that require aircrews to decide more frequently under time pressure. This shift is exemplified by the deployment of Autonomous Collaborative Platforms (ACPs), i.e., AI-enabled teamed drones operating alongside crewed fighters and proposing actions such as target validation. This study examines whether the processing of target-validation situations proposed by ACPs can be improved through cognitive training. Forty military aircrew members participated in a virtual reality flight mission and were assigned to either a control or a training group. Both groups completed identical pretest and posttest separated by 24 hours, while only the training group performed a 45-minute computer-based training session focused on repeated target-validation tasks. Results show that training significantly reduced error rates and response times, in line with theories of expertise acquisition. In contrast, perceived workload decreased similarly in both groups, suggesting a dissociation between objective performance gains and subjective workload. These findings support the relevance of cognitive training for preparing aircrews for future collaborative combat involving ACPs.
Beyond-visual-range (BVR) air combat is a challenging reinforcement-learning domain characterized by partial observability, long-horizon decision making, energy management, and limited weapons. We present BVR Sim, an open-source Gymnasium-style environment designed for heterogeneous air-combat reinforcement learning. BVR Sim supports multiple JSBSim aircraft models, including the F-15, F-16, F/A-18, and F-22, with configurable weapons, sensors, controllers, and opponents. A unified tactical action interface specifies desired heading, altitude, speed, and weapon release above aircraft-specific inner-loop controllers, enabling policies to operate across heterogeneous platforms. The environment provides interchangeable Python and accelerated C++ backends, entity-oriented observations, compositional rewards, scripted opponents, replay and visualization, and adapters for multi-agent learning frameworks. At a 0.4-s decision interval, the C++ backend achieves 104 simulated seconds per wall-clock second in 1-vs-1 and remains practical through 10-vs-10 scenarios. A policy trained only on the F-16 transfers without retraining to four unseen aircraft, reaching a 45.5% mean win rate with aircraft-specific controller adaptation. MAPPO and HAPPO experiments further verify end-to-end compatibility with standard multi-agent reinforcement-learning pipelines.
Controlling multiple autonomous, uncrewed aircraft, or autonomous collaborative platforms (ACPs), during flight missions is a difficult multitasking effort. Due to the complexities and risks related to aircraft operations, it is imperative that the designs of emerging ACP management controls yield as few errors as possible during such missions. One way to achieve this is to examine, classify, and enumerate the types of errors that pilots make when flying missions with autonomous teammates. Here, we examine performance in a simulated piloting task in which individuals must direct four ACPs to eliminate ground targets while simultaneously piloting their own aircraft. We conducted structured failure analyses on different behavioral outcomes, integrating two different human factors taxonomies. The results of these analyses indicate that task failures were jointly attributable to cognitive capacity limits, configuration complexity, and ACP interface design.
Hailey Ramos, Taylor Curley, Megan B. Morris et al.· Proceedings of the Human Fac...· 0 citations
It is argued that agentic AI should be approached as a socio-technical design problem, where interfaces, oversight mechanisms, and evaluation practices are as critical as algorithms.
Timothy Merritt, Alejandro Jarabo-Peñas, Juan Bravo-Arrabal et al.· 0 citations
The aim of air traffic control is to guide aircraft from its departure to its destination in a safe and efficient manner. Keeping the aircraft separated in accordance with the regulations is one of the most crucial tasks, which in dense traffic often induces a high workload for the air traffic controller. Automated support systems for conflict detection and resolution would be a valuable addition to the air traffic control systems and has been an active area of research for decades. Reinforcement learning has recently been suggested as a promising alternative to traditional algorithms. A key challenge in air traffic control is that there is a significant delay between an action and its execution, which can have a significant negative impact in reinforcement learning. In this paper we show that by adding data from the trajectory prediction system to the observation space in combination with larger networks and reward sharing, we can overcome the problems introduced by the action delays. We also present a novel method of creating a low-cost ensemble by combining the agents from one training session and using the critics value function to vote on the actions to further improve our results.
Jens Nilsson, Jonas Unger, Gabriel Eilertsen· IEEE Open Journal of Intelli...· 0 citations
This paper examines the current applications and emerging development trends of
unmanned aerial vehicles (UAVs), with emphasis on the technological and operational evolution
of contemporary drone systems across civil and military domains. The study reviews the
expanding use of UAVs in precision agriculture, infrastructure inspection, logistics, emergency
response, urban air mobility, intelligence, surveillance and reconnaissance (ISR), electronic
warfare and tactical operations, highlighting the operational advantages offered by unmanned
platforms in diverse mission environments.
The principal challenges affecting future UAV deployment are also analyzed, including
battery endurance, cybersecurity, communication reliability, regulatory compliance, airspace
integration and environmental constraints. Attention is given also to the increasing role of
artificial intelligence, autonomous mission execution, swarm coordination, advanced propulsion
technologies and digital airspace management in shaping the next generation of unmanned
aircraft systems.
Current research directions indicate a transition from individually operated drones toward
interconnected, intelligent and highly autonomous aerial systems capable of cooperative decision-
making and large-scale deployment. The analysis shows that future progress in unmanned
aviation will depend not only on improvements in aircraft performance but also on advances in
systems integration, secure communications, resilient autonomy and regulatory frameworks that
enable the safe and efficient operation of increasingly complex UAV ecosystems.
Andrei Bencze· SCIENTIFIC RESEARCH AND EDUC...· 0 citations
This article surveys how simulation has been used to study special operations in non-segregated airspace, where civil, military and unmanned aircraft increasingly share the same volume. The operations of interest include the handling of non-cooperative unmanned aircraft, counter-UAS activity, military flights transiting civil airspace, and cooperative maneuvers such as air-to-air refueling. How this mixed traffic behaves, the workload it imposes on controllers, and the value of any new procedure or decision-aiding tool are empirical questions that cannot be answered safely, affordably, or repeatably in live flight. The discussion is organized around four threads: the role of simulation as a research instrument; the fast-time and real-time human-in-the-loop paradigms, together with the platforms that embody them (FACET, BlueSky, ESCAPE, en-route digital twins); the conflict-detection, conflict-resolution and detect-and-avoid methods that have been developed and validated computationally; and the human-factors evidence from controller-in-the-loop experiments. Drawing these threads together, the article argues that a dedicated, high-fidelity simulation capability is not a convenience but a precondition for safely integrating special operations into shared airspace, and it identifies where the literature remains thin, most notably the scarcity of platforms purpose-built for special operations and for non-cooperative traffic.
Ana-Maria Bordei, C. Nae· Romanian Journal of Technica...· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
