Landscape-Aware Bandit Hyper-Heuristics for Online Operator Selection in UAV Inspection Routing

2026-05-14

AI Summary

This work addresses the online operator selection problem in multi-depot drone inspection by proposing a landscape-aware LA-BHH method. Integrating static map features with dynamic search states, LA-BHH employs a contextual multi-armed bandit controller based on LinUCB to adaptively select among local search operators—including 2-opt, swap, relocate, and Or-opt—during optimization. The approach further incorporates a 2-opt repair mechanism and a stagnation-aware strategy to enhance solution quality. Evaluated on 45 Euclidean TSP instances, LA-BHH achieves an average final gap of 0.0223 and a convergence AUC of 0.0389, reducing the final gap by 17.6%, 22.6%, and 68.2% compared to UCB-HH, random selection, and nearest-neighbor construction, respectively, thereby significantly improving both solution quality and computational efficiency.

Abstract

UAV multi-site inspection often reduces to choosing a high-quality visiting order after target sites have been extracted from a map. This paper develops LA-BHH, a landscape-aware bandit hyper-heuristic that learns an operator-selection policy online for this routing layer. LA-BHH treats 2-opt, swap, relocate, and Or-opt moves as low-level arms, builds context from static landscape descriptors and online search-state features, and updates a LinUCB controller from improvement rewards during the same run. Experimental results on 45 generated Euclidean TSP instances show that LA-BHH achieves the best mean final gap and convergence AUC, with 0.0223 and 0.0389 respectively. It reduces final gap by 17.6\% over UCB-HH, 22.6\% over Random-HH, and 68.2\% over nearest-neighbor construction. Ablation results further show that contextual credit assignment, 2-opt repair, and stagnation-aware state use are the main contributors.

Problem

• UAV inspection routing
• online operator selection
• hyper-heuristics
• combinatorial optimization
• traveling salesman problem

Innovation

• landscape-aware
• bandit hyper-heuristic
• online operator selection
• LinUCB
• UAV routing