Advanced dynamics modeling and innovative estimation and control strategies for air delivery applications

This dissertation investigates estimation and control strategies for cooperative aerial transportation with multirotor vehicles carrying cable-suspended payloads. The central challenge is the dynamic coupling introduced by the suspended mass, which renders the system underactuated, highly nonlinear, and prone to oscillations that degrade safety and performance. A high-fidelity simulator based on first-principles models of single and multi-vehicle slung-load systems is developed, serving as a design and verification platform under model uncertainty, exogenous disturbances, and sensor noise. Three main technical advances are contributed. (i) A compact nonlinear formation controller derived from an Eulerian model exhibits two-time-scale behavior: fast modes stabilize formation geometry and damp payload swing, while slow modes achieve trajectory tracking. Stability and gain selection are guided by singular perturbation arguments and Lyapunov criteria. (ii) A baseline controller from a Lagrangian formalization introduces a physically insightful parameterization of payload swing, enabling simultaneous formation keeping, trajectory tracking, and oscillation suppression in a single feedback law. (iii) A robustness layer based on L1 adaptive control augments the baseline while preserving its nominal properties, compensating matched and unmatched uncertainties, external disturbances, and actuator faults without compromising closed-loop stability. Complementing control, a discrete-time estimator requiring only standard onboard inertial measurements is proposed. Using a Fading Gaussian Deterministic Filter, the method estimates payload swing angles and rates without extra sensors. The framework is further extended to a single multirotor transporting two serially-connected payloads, enabling multi-parcel delivery within a single flight. All methods are validated through high-fidelity simulations and experimental campaigns, demonstrating reliable swing attenuation, accurate trajectory tracking, and robustness to disturbances and parameter uncertainty on resource-constrained avionics. These contributions yield a unified framework for aerial mobility and delivery scenarios, applicable to logistics, search-and-rescue, and disaster-relief operations.