DA-NBV: A Direction-Aware Next-Best-View Planner for Efficient 3D Reconstruction of Ships at Sea

2026-08-08

AI Summary

This study addresses the challenge of incomplete 3D reconstruction of maritime vessels, which arises from their complex structures, self-occlusions, and wave-induced motion that hinder traditional methods. To overcome this, the authors propose a direction-aware Next-Best-View (NBV) planner that leverages a learnable Position Advantage Field (PAF) to model observation history, integrates a locally constrained action space, and employs a nonlinear coverage-based reward mechanism to guide an autonomous drone toward optimal viewpoints. The work introduces SeaShip-3D, the first vessel-centric 3D dataset, alongside a realistic sea-condition simulation environment. Experimental results demonstrate that the proposed approach improves reconstruction completeness by approximately 3 percentage points, reduces Chamfer distance by 43%, and significantly enhances path efficiency across varying sea states.

Abstract

Accurate 3D reconstruction of ships at sea is important for maritime supervision, damage assessment, and autonomous maritime operations. Although 3D reconstruction has advanced considerably, high-quality data acquisition still largely relies on manually designed trajectories or skilled operators, resulting in high costs and limited scalability. Next-best-view (NBV) planning automates this process by selecting subsequent viewpoints based on the current state. However, existing NBV policies mainly model spatial occupancy while overlooking directional observation history. This limitation is particularly problematic for ships: their complex superstructures and severe self-occlusions require observations from multiple viewpoints, and insufficient directional coverage often yields incomplete reconstructions. These challenges are further amplified at sea, where wave-induced heave, roll, and pitch continuously alter the ship's pose and surface visibility. Meanwhile, wind disturbances and limited onboard power impose stricter requirements on scanning efficiency. To address these challenges, we propose DA-NBV, a direction-aware NBV policy that augments the conventional occupancy state with directional observation statistics. We introduce a learnable Position Advantage Field (PAF) that uses directional information to guide viewpoint selection. The policy further adopts a locally constrained action space and a nonlinear coverage-shaping reward to improve scanning efficiency. We also develop the ship-oriented SeaShip-3D dataset and a configurable sea-state simulation environment. Experiments under varying heave, roll, and pitch conditions show that DA-NBV improves reconstruction completeness by approximately 3 percentage points and reduces Chamfer distance by 43% while achieving higher path efficiency.

Problem

• Next-Best-View
• 3D reconstruction
• directional observation
• ship at sea
• self-occlusion

Innovation

• direction-aware
• next-best-view planning
• Position Advantage Field
• 3D reconstruction
• maritime simulation