Underwater robotics · Archaeological mapping
Development of Experimental and Exploratory Photogrammetry for Supporting Underwater Archaeological Studies using Remotely Operated Vehicle
Development of a low-cost ROV and multi-modal mapping framework for underwater archaeological documentation in Indonesia.
01 · Overview
DEEP-UNDER ROV develops a low-cost remotely operated vehicle and multi-modal mapping methods for underwater archaeological documentation in shallow tropical waters.
The research covers mechanical and electrical design, vehicle control, state estimation, underwater imaging, visual-inertial SLAM, photogrammetry, acoustic sensing, and 3D reconstruction.
Long-term direction
The long-term objective is a multi-modal survey platform for digital-twin generation, autonomous data acquisition, and coordinated multi-vehicle mapping.
Upcoming field experiment
The current platform will be used for field testing, underwater data collection, positioning, and mapping experiments.
02 · Research focus
The 2025 work evaluated a low-cost ROV assembled from components available through the domestic supply chain. Underwater cameras and pressure sensing were integrated with the mechanical and electrical systems. The platform and an initial visual-SLAM pipeline were tested in controlled and coastal environments.
The tests evaluated the effects of illumination, turbidity, vehicle motion, viewing geometry, and sensor integration.
The 2026 work focuses on data-acquisition reliability and geometrically consistent reconstruction.
Hardware Mechanical design, waterproofing, power, leak detection, and six-degree-of-freedom actuation
Control Modelling, simulation, visual servoing, trajectory tracking, and active perception
Estimation & SLAM Stereo vision, inertial sensing, depth, acoustics, Kalman estimation, and underwater camera modelling
3D reconstruction Photogrammetry, point clouds, Gaussian splatting, acoustic constraints, and quality assessment
03 · Progress
01
The first four-degree-of-freedom prototype established the mechanical frame, propulsion arrangement, onboard electronics, cameras, pressure sensing, and tethered operator workflow. Laboratory and pool tests confirmed that the platform could support underwater inspection and visual-mapping research.
02
Field tests at Pramuka Island evaluated underwater operation at depths of up to 12 metres. The recorded imagery was used for camera-trajectory estimation and sparse 3D reconstruction.
03
The second four-degree-of-freedom iteration refined the enclosure, sealing, component layout, and electrical integration, moving the project from an initial feasibility platform toward a system designed for repeated experiments.
04
The current platform expands motion capability to six degrees of freedom and improves sealing, cable management, power switching, leak detection, fibre-optic communication, and ground-control integration.
05
This stage covers a Gazebo and ROS 2 simulator, stereo-enhanced image-based visual servoing, and visual-inertial Kalman estimation. A PINAX underwater camera model was integrated into a stereo visual-inertial SLAM pipeline and evaluated through pool tests at Saraga.
Visual-SLAM and Gaussian-splatting experiments use data from Saraga Pool and Pramuka Island, including data acquired around an ADCP support frame.
06
The 21–26 October field experiment will test the platform and mapping workflow at Pramuka Island. The programme includes underwater data collection, positioning and reconstruction experiments, and preparation of the final 2026 research outputs.
04 · People and partners
Current
Graduated
Source of funds
Riset ITB 2025 and Riset Dosen Muda ITB 2026
05 · Documentation
06 · Publications