Hybrid discrete continuous optimization for connectivity-constrained UAV path planning

Unmanned aerial vehicle (UAV) missions depend on reliable connectivity while operating within cellular and wireless communication environments. Most existing communication-aware planning methods rely on instantaneous QoS (Quality of Service) limitations or heuristic penalties. These don’t guarantee reliability across the entire mission. This paper formulates connectivity-constrained UAV path planning as an explicit cumulative outage-budget-constrained optimization problem and implements it using a hybrid discrete-continuous planning framework. The proposed formulation first computes a feasible path on a nominal connectivity map under an explicit outage budget and subsequently refines it into a smooth trajectory while preserving planning-level outage-budget compliance through sampled feasibility evaluation. Theoretical analysis establishes feasibility conditions and shows that outage compliance can be preserved during refinement under suitable connectivity margin conditions. Representative simulations under stochastic channel realizations ( \(N=50\) Monte Carlo trials) demonstrate that the proposed framework achieves near-zero empirical outage violations across Monte Carlo realizations while maintaining path length within approximately 1–3% of the distance-optimal baseline. Compared with penalty-based planners, the proposed method reduces outage violation rate from over 45% and exhibits statistically significant improvements ( \(p<10^{-3}\) ) with large effect sizes. A preliminary multi-UAV extension illustrates how the proposed planning formulation can be generalized to coordinated routing under the same planning assumptions. These results demonstrate that embedding an explicit cumulative outage budget into the planning formulation provides a scalable, interpretable, and reliability-aware framework for connectivity-constrained UAV path planning under representative communication-aware planning scenarios.