The research, led by Yiran Wei and Ming He of the Army Engineering University of the PLA in Nanjing, together with colleagues at the Beijing Institute of Technology and the National Key Laboratory on Near-Surface Detection, tackles two goals that usually pull against each other: rapid response and stability maintenance. A formation that reacts quickly to command changes tends to oscillate, while a formation tuned for smoothness tends to lag. The team’s answer is a decentralized adaptive tracking control method built on a leader-follower coordination strategy, in which each aircraft in the cluster is responsible for its own tracking behavior while the group as a whole follows a virtual leader.

The virtual leader idea is central to how the scheme scales. Rather than designating one physical drone as the boss whose failure would doom the mission, the researchers define a mathematical reference trajectory that every aircraft in the cluster tries to track, offset by its assigned position in the formation. Each CRA then adjusts its relative position to its neighbors to hold the overall shape. Because the coordination is decentralized, no single aircraft has to broadcast commands to the whole swarm, and the control burden does not grow explosively as the cluster expands. This design philosophy echoes a broad body of work on multi-agent formation control, from graph-based model predictive approaches to flocking-inspired swarm navigation, but the new study adapts it specifically to the dynamics of coaxial rotors.

To get there, the authors first built a mathematical model of a single CRA and then extended it to the cluster system. The model separates the aircraft’s motion into two subsystems with distinct jobs. The position subsystem maintains the formation by adjusting where each aircraft sits relative to the others, while the attitude subsystem is responsible for driving each aircraft’s orientation to a desired stable state. This split matters because coaxial aircraft are underactuated: they have fewer independent control inputs than degrees of freedom, so position is controlled indirectly through attitude. Getting the attitude loop right is therefore a precondition for getting the position loop right.

For the position subsystem, the team turned to back-stepping control, a well-established nonlinear design method that works backward through the layers of a system’s dynamics. In a back-stepping design, the controller treats some of the internal states of the aircraft, such as velocity components, as virtual controls for the outer states, such as position. It designs a stabilizing law for the outer loop, computes what the virtual controls would need to be to achieve it, and then steps inward, designing control laws for each successive layer until it reaches the true physical inputs. The result is a controller that respects the nonlinear coupling in the aircraft’s equations of motion rather than approximating them with a linearized model, which is why back-stepping has become a favorite tool for quadrotors and, increasingly, for coaxial platforms.

The attitude tracking controller is where the machine learning enters. Coaxial rotors generate complicated aerodynamic interactions, including blade-vortex interactions between the upper and lower rotors, that are difficult to model precisely. Rather than trying to capture every disturbance term analytically, the researchers designed an adaptive radial basis function neural network, or RBFNN, that learns the unknown nonlinearities online. An RBFNN is a particular class of neural network in which each neuron responds to inputs within a localized region of the input space, using a radial basis function such as a Gaussian centered at a particular point. This locality gives RBFNNs a powerful theoretical property: under mild assumptions, they can approximate any continuous nonlinear function over a compact domain. In the control context, that means the network can represent the lumped uncertainty in the attitude dynamics, whatever its exact form, and the adaptive tuning laws adjust the network weights in real time as the aircraft flies.

Critically, the team did not simply trust the neural network to behave. The stability of both the position and attitude controllers is mathematically proven using Lyapunov theory, the standard framework for demonstrating that a controlled system’s energy-like function decreases over time, driving the tracking errors to zero or to a bounded neighborhood of zero. The Lyapunov analysis ties the neural network’s weight adaptation to the error dynamics, ensuring that the learning process itself cannot destabilize the aircraft. This combination of learning-based approximation with rigorous stability proof is what distinguishes the approach from purely empirical machine learning controllers, and it addresses one of the biggest obstacles to deploying neural networks on real flight hardware.

Verification proceeded in two stages. First, the team ran numerical simulations of intelligent tracking scenarios, and then they moved to software-in-the-loop, or SITL, simulations, in which the flight control code runs on a simulated autopilot environment closer to real deployment conditions. SITL testing is a standard bridge between pure simulation and flight tests because it exposes the controller to computational delays, discretization effects, and software architecture issues that idealized numerical models ignore. Across both verification stages, the reported trajectories of the CRAs were smooth and continuous, with excellent position response and minimal amplitude vibration, the two qualities the study set out to reconcile.

The choice of coaxial platforms for cluster flight is not arbitrary. Compared with conventional multirotors, coaxial aircraft offer higher thrust-to-footprint ratios and better tolerance to rotor damage, and recent work has explored their use in everything from micro-scale twin-rotor systems with internet-of-things control to coaxial tilt-rotor designs with finite-time and adaptive attitude controllers. Earlier studies have applied back-stepping sliding mode control to coaxial trajectory tracking, fuzzy logic methods with time-delay estimation, dynamic inversion for compound coaxial helicopters, and adaptive neural network approaches for coaxial micro aerial vehicles. The new study’s contribution is to bring these threads together at the cluster level, pairing the back-stepping structure that has proven effective for single-aircraft tracking with neural network adaptation and a decentralized formation architecture.

The implications reach beyond any single aircraft type. Formation tracking of aerial swarms underpins applications ranging from perimeter surveillance and cooperative sensing to search-and-rescue and coordinated defense missions, and the field has been moving steadily toward controllers that can handle unmodeled dynamics, switching communication topologies, and uncertain environments. By demonstrating that an RBFNN-augmented back-stepping controller can deliver both fast formation response and stable flight for a coaxial cluster, validated through Lyapunov-guaranteed stability analysis and software-in-the-loop testing, the researchers offer a template that other swarm platforms could adapt. The work was supported in part by the National Natural Science Foundation of China, the National Key Research and Development Program of China, and the National Key Laboratory on Near-Surface Detection, and the authors report no conflicts of interest. The next step for this line of research, as with all simulation-validated control schemes, will be demonstrating the same smooth, vibration-free tracking on physical aircraft flying in the open air.

Subject of Research: Decentralized adaptive formation tracking control of coaxial-rotor aircraft clusters using back-stepping and RBF neural networks

Article Title: Intelligent tracking control of CRA cluster via back-stepping and RBFNN strategy

Article References: Wei, Y., He, M., Pan, Z., Wu, H., Liu, C., & Jing, P. (2026). Intelligent tracking control of CRA cluster via back-stepping and RBFNN strategy. International Journal of Machine Learning and Cybernetics, 17(9), Article 445. https://doi.org/10.1007/s13042-026-03215-0

Image Credits: AI Generated

DOI: 10.1007/s13042-026-03215-0

Keywords: coaxial rotor aircraft, formation control, back-stepping control, RBF neural network, adaptive control, leader-follower strategy, Lyapunov stability, UAV swarm, software-in-the-loop simulation, flight control, intelligent tracking, multi-agent systems

The research, published on 30 July 2026, addresses two intertwined challenges that have long frustrated engineers working on coordinated flight. The first is geometric: how does a group of aircraft maintain a coherent shape when the environment is littered with obstacles of varying size? The second is operational resilience: in combat scenarios, where reliability and fault tolerance are not optional extras but survival requirements, how can a formation continue executing its mission after some of its members are damaged or destroyed? The team’s answer draws on a control philosophy known as behavior-based control, an approach with deep roots in robotics that treats each agent as the simultaneous host of several competing, simple behaviors whose combined output produces complex, adaptive group motion.

Behavior-based formation control traces its lineage to a landmark 1998 study by Tucker Balch and Ronald Arkin, who showed that multirobot teams could hold formations by blending elementary behavioral urges such as move-to-goal, avoid-obstacle, and maintain-formation. Rather than computing a single globally optimal trajectory for the whole group, which becomes computationally intractable as the number of agents grows, each robot weighs local priorities and acts. The Qingdao team revives and extends this philosophy for aerial swarms, where the stakes are higher because a mid-air split of a formation can strand aircraft on the wrong side of a structure with no coherent plan for rejoining.

The first novel behavior the authors introduce is wing contraction. In dense obstacle environments, the greatest threat to a formation is not collision with any single object but segmentation: the group flowing around an obstacle like a stream around a boulder, splitting into two disconnected halves that may never reorganize. The wing-contraction behavior counters this by temporarily narrowing the formation’s lateral extent, compressing the group into a tighter envelope that can slip through gaps between obstacles without being severed. Once the constriction is passed, the formation expands back to its nominal geometry. The technique is conceptually simple but addresses a failure mode that more elaborate trajectory planners often handle poorly, because it preserves the formation’s integrity as a primary objective rather than treating each aircraft’s path as an independent optimization problem.

Contraction alone, however, cannot solve every encounter. When the swarm confronts an obstacle too large to squeeze past, the formation must go around it, and doing so in a coordinated way is far harder than steering a single aircraft around a wall. For this case the researchers designed a wall-following behavior that guides the entire formation along the edges of large obstacles, keeping the group coherent as it circumnavigates the blockage. Wall following is a classic technique from mobile robotics, where ground robots trace the perimeter of objects using proximity sensors, but transplanting it to a multi-UAV formation requires the behavior to act on the group’s shared reference frame rather than on any individual aircraft. The formation effectively behaves as a single elastic body that slides along the obstacle’s boundary, its members maintaining relative positions while the collective path bends around the obstruction.

The third and arguably most consequential contribution is the formation reconfiguration strategy, designed explicitly for the reliability and fault-tolerance demands of combat scenarios. When a UAV is lost, whether through enemy fire, mechanical failure, or a hard collision, the remaining formation must not simply carry a permanent gap. The proposed strategy enables autonomous maintenance and reconstruction: surviving aircraft detect the loss, recompute their target positions, and redistribute themselves to close the hole, restoring a functional formation shape with fewer members. This allows the mission to continue even after attrition. The authors describe the strategy as ingenious in its economy, relying on the same behavioral architecture rather than requiring a separate, heavyweight replanning system to be activated mid-flight, which is often where conventional approaches lose precious seconds or fail outright when communication links are degraded.

What distinguishes this work from much of the existing literature on formation reconfiguration is the emphasis on continuity under attack. Recent surveys of UAV swarm formation control, including comprehensive reviews published in Drones and Progress in Aerospace Sciences, catalog a rich toolbox of reconfiguration methods: hybrid particle swarm and genetic algorithms, modified artificial bee colony optimization, pigeon-inspired optimization for resilience, and pseudospectral trajectory optimization. These methods can produce elegant new formations but typically assume a calm moment in which the swarm can pause, replan, and execute. The Qingdao strategy instead embeds reconfiguration into the ongoing flight behavior, so that the transition happens while the mission proceeds, an essential property when the loss of a UAV is caused by an adversary who may strike again.

The authors validated their approach through simulation experiments in two typical environments, chosen to exercise the full behavioral repertoire. The simulations demonstrated that the wing-contraction behavior prevents segmentation in cluttered passages, that the wall-following behavior successfully guides formations around large obstructions, and that the reconfiguration strategy restores formation integrity after individual UAV losses. The paper reports that these experiments confirm the effectiveness of the proposed control methods and strategies, providing a proof of concept for a system in which the same underlying behavioral rules handle navigation, obstacle negotiation, and damage recovery without mode-switching between fundamentally different control regimes.

The broader significance of the work lies in its timing. UAV swarms are moving from research demonstrations toward operational deployment in surveillance, disaster response, infrastructure inspection, and military operations, and the field’s state-of-the-art reviews consistently identify robustness in complex, adversarial environments as a defining research challenge. Formations that shatter on contact with complexity are of limited use. By combining classical behavior-based control, which is computationally light and inherently distributed, with targeted behaviors for the two most common failure scenarios, segmentation and attrition, the Qingdao team offers a template for swarms that degrade gracefully rather than catastrophically. The approach also sidesteps the heavy communication and computation burdens of centralized planners, since each UAV’s decisions emerge from locally evaluated behaviors.

There remain, of course, the usual caveats that separate simulation from the sky. Real flight adds wind gusts, sensor noise, communication latency, and the nonholonomic flight dynamics of fixed-wing and rotorcraft platforms, all of which stress-test behavioral arbitration schemes in ways that idealized simulators do not. The authors note that the simulation code may be made available on request, and the work was carried out without external funding. Still, the study’s core insight is likely to endure: a swarm does not need to be smarter to be tougher. It needs a small set of well-chosen behaviors, including the instinct to pull in its wings in tight spaces, to hug the wall when the way is blocked, and to close ranks when a comrade falls. In an era when drone formations are expected to operate where things go wrong, those instincts may prove to be the difference between a mission accomplished and a mission lost.

Subject of Research: Behavior-based cooperative control and fault-tolerant formation reconfiguration for multi-UAV swarms in obstacle-rich environments

Article Title: Behavior-Based Multi-UAV Formation Control Under Random Attacks

Article References: Behavior-Based Multi-UAV Formation Control Under Random Attacks. (n.d.). https://doi.org/10.1007/s42405-026-01274-9

Image Credits: AI Generated

DOI: 10.1007/s42405-026-01274-9

Keywords: UAV swarm, formation control, behavior-based control, formation reconfiguration, fault tolerance, obstacle avoidance, wall following, multi-agent systems, cooperative control, aerospace engineering, drone resilience, autonomous navigation