Learning-Accelerated Optimization-based Trajectory Planning for Cooperative Aerial-Ground Handover Missions

TL;DR

This paper introduces a neural network-based surrogate planner that accelerates trajectory optimization for cooperative UAV-UGV missions, achieving over three times faster solutions with guaranteed feasibility.

Contribution

It proposes a learning-augmented framework using LSTM networks to provide warm starts for centralized optimization, significantly improving speed and success rate.

Findings

Over threefold speedup in trajectory planning.

100% success rate in optimization with warm starts.

Effective combination of data-driven inference and model-based refinement.

Abstract

This paper presents a learning-augmented trajectory planning framework for cooperative unmanned aerial vehicle (UAV) and unmanned ground vehicle (UGV) handover missions. While centralized trajectory optimization ensures dynamic feasibility and task optimality, its high computational cost limits real-time applicability. We propose a neural surrogate planner utilizing decoupled encoder-decoder long short-term memory (LSTM) networks to generate coordinated handover trajectory predictions from the task specifications. These predictions serve as informed warm starts for the downstream centralized optimizer, thereby accelerating convergence to dynamically feasible solutions. Benchmark evaluations demonstrate that the learning-augmented planning framework achieves more than a threefold speedup and 100% optimization success rate compared to cold start optimization. The results indicate that combining data-driven inference with model-based refinement enables fast and reliable trajectory generation for heterogeneous multi-robot systems.