Aug 2026· Science Advances· Vol 12· 0 citations· 52 references
Medicine
TL;DR
A portable and automated drone-assisted multicamera 3D tracking framework for georeferenced trajectory reconstruction and establishes DAM3T as a scalable measurement platform that extends outdoor tracking from local-coordinate reconstruction to georeferenced measurement of complex dynamic systems.
Abstract
Accurate measurement of three-dimensional (3D) trajectories in outdoor environments is essential for studying complex dynamic processes such as atmospheric transport, aerial vehicle motion, and collective behavior. However, obtaining quantitative trajectory measurements over large outdoor volumes remains challenging because multicamera systems require reliable calibration and synchronization under field conditions. Here, we introduce a portable and automated drone-assisted multicamera 3D tracking (DAM3T) framework for georeferenced trajectory reconstruction. The system estimates multicamera extrinsic parameters from an autonomous drone flight without dedicated calibration markers and aligns the reconstructed coordinate system with geographic coordinates (WGS84) using real-time kinematic Global Positioning System (RTK-GPS) measurements. Validation under indoor motion-capture and outdoor RTK-GPS conditions demonstrates trajectory errors below 2%. We further demonstrate the framework through Lagrangian analysis using passive tracers, georeferenced tracking of multiple unmanned aerial vehicles, and 3D motion analysis in sports scenes. This establishes DAM3T as a scalable measurement platform that extends outdoor tracking from local-coordinate reconstruction to georeferenced measurement of complex dynamic systems.
Accurate velocity estimation in GPS-denied environments remains a core challenge for autonomous unmanned aerial vehicle (UAV) navigation. Our previous work demonstrated that fusing optical flow (OF) with dead reckoning (DR) substantially reduces position drift compared to inertial-only solutions. However, velocity estimates derived from dense optical flow are degraded by two systematic effects: (1) incorrect metric scaling when the camera footprint covers heterogeneous terrain types—particularly at forest–field transitions where the visible surface elevation differs significantly from bare-ground elevation; and (2) flow magnitude bias introduced by scene texture and structural properties. This paper presents three targeted improvements to a software pipeline for optical flow-assisted UAV navigation. First, single-point above-ground-level (AGL) estimation is replaced by camera footprint area mean sampling over co-registered Digital Terrain Model (DTM) and Digital Surface Model (DSM) rasters, with the surface model adopted consistently for OF metric scaling. Second, a one-dimensional Kalman filter with an innovation gate suppresses velocity spikes caused by abrupt terrain transitions. Third, a compact data-driven correction module uses selected flow, texture, and terrain descriptors to estimate a multiplicative velocity correction factor aligned with GPS-derived reference speed available during calibration and offline evaluation but not required during GPS-denied operation. Experiments on three real PX4-logged flight missions (Log 258 for calibration and Logs 259–260 for independent evaluation) totalling 7.3 min show terrain-dependent behaviour. On the mixed forest–field validation flight (Log 260), the improved pipeline reduces velocity mean absolute error (MAE) by 71% (1.04 m/s → 0.30 m/s) and dead-reckoning position MAE by 88% (59.8 m → 7.1 m), compared to the baseline from our previous work. On a flat open-terrain validation flight (Log 259), the terrain-aware modifications leave the terrain-insensitive baseline essentially unchanged, providing a control case for the proposed DSM-based scaling mechanism.
Jakub Walczak, Piotr Targowski, Szymon Chmielewski et al.· Italian National Conference...· 0 citations
Robotic platforms operating in GPS-denied environments require robust ego-motion estimation systems that fuse complementary sensor modalities under onboard computational constraints. This paper proposes a navigation framework estimating six-degree-of-freedom (6 DoF) robot pose in unstructured scenes using a monocular camera stream, inertial measurement unit (IMU) data, and sparse depth cues within the multi-state constraint Kalman filter (MSCKF) architecture. The key innovation integrates 3D landmark measurements into visual feature tracks, reducing positional uncertainty and drift accumulation compared to vision-only approaches. The method is efficient enough for resource-constrained systems such as micro aerial vehicles and small ground robots. The measurement fusion strategy is analytically derived and evaluated on aerial robot trajectory datasets. Results show improved tracking accuracy and stability in challenging indoor and outdoor scenarios without GPS, enabling prolonged autonomous missions in complex 3D environments with real-time pose feedback and low computational burden.
Pratik Dhameliya· 2026 6th International Confe...· 0 citations
Recently, unmanned aerial vehicle (UAV)-borne radar (radio detection and ranging) calibration system has been found to be promising in replacing the conventional balloon-based calibration system which incurs challenges of deciding the reference location. However, in the UAV-borne calibration system major challenges arise due to the inherent vibration of the drone and its movement due to wind during the flight. Therefore, estimation of such vibrational noise to compensate such effects is crucial to improve the localization accuracy in weather radar calibration. The available global postioning system (GPS) positioning systems are incapable of capturing such positional movements of the UAV up to the submillimeter level. Available filtering algorithms can estimate such UAV vibration using an inertial measurement unit (IMU) but does not provide a reference location of the calibrator on the UAV unless GPS satellite data is used. In this work, a complete system is proposed that estimates the positional displacement of UAV from the GPS coordinates and provides a reference position with 0.604-mm spatial resolution utilizing IMU data. The proposed system is cost-effective as it utilizes low-cost components like Raspberry Pi, NEO 6M GPS receiver, and MPU 6050 IMU sensor. The total weight of the complete system is 158 g without battery, and dimension is $14\times 10.5\times 4.0$ cm, which is UAV payload friendly. Experimental results are presented to validate the research work with static and controlled vibration in the laboratory as well as outdoor environment during UAV flight. Effectiveness of the filtering mechanism in estimating the vibration and reducing its effect during localization of UAV-aided radar calibrator on flight is also presented.
Shivam Saini, Ayandip Garai, Bhushan Siddhu Maghade et al.· IEEE Transactions on Instrum...· 0 citations
This paper presents a three-dimensional (3D) unmanned aerial vehicle (UAV) localization framework based on optical round-trip time (RTT) ranging and multilateration. Multiple fixed optical transmitters estimate distances via laser reflections from a UAV equipped with passive retroreflectors, enabling infrastructure-assisted localization without active onboard hardware or additional energy consumption. The system model incorporates additive Gaussian ranging noise, and the localization problem is formulated as a non-linear least-squares estimation problem. Monte Carlo simulations demonstrate a median localization error of approximately 0.055 m, with 90% and 95% of errors below 0.11 m and 0.14 m, respectively, under centimeter-level ranging noise. The results show that localization accuracy depends on transmitter geometry and UAV altitude, with errors increasing at higher altitudes and under larger noise levels. Furthermore, increasing the number of transmitters improves accuracy through enhanced measurement redundancy. The results provide practical design guidelines for optical UAV tracking systems.
Khadijeh Ali Mahmoodi, Bastien Béchadergue, Luc Chassagne et al.· International Symposium on C...· 0 citations
Accurate three-dimensional (3D) localization and trajectory generation of key objects in complex environments remain challenging due to limitations of existing single- or multi-modal methods, such as low accuracy, slow processing speed, and sensitivity to occlusions, especially when relying on single-modality sensing. This paper proposes LCG-3D, a novel framework for local key object 3D positioning and trajectory generation under cross-modal geometric consistency constraints. By integrating heterogeneous sensor data, including RGB images, depth maps, and LiDAR point clouds, LCG-3D enforces local geometric consistency to align multi-modal observations in 3D space. The algorithm selectively focuses on key objects, reducing computational overhead while improving robustness in dynamic or occluded environments. A trajectory generation module further predicts object motion by leveraging both current localization and historical geometric patterns. Extensive experiments on publicly available multi-modal datasets demonstrate that LCG-3D achieves superior localization accuracy and trajectory fidelity compared with state-of-the-art methods, highlighting its potential for applications in intelligent transportation, robotics, and augmented reality.
Pengfei Zhang, Mini Wu, Xiaojian Zhong et al.· International journal of pat...· 0 citations
To bypass the high computational overhead and environment-specific mapping dependencies of traditional indoor localization, this work introduces a cost-effective navigation framework that enables standard autopilot controllers to operate indoors via dynamic GPS retransmission. By integrating OAK-D AI cameras for wide-area target detection with Software-Defined Radio (SDR) technology, the system generates real-time, localized GPS signals to provide seamless position inputs to commercial off-the-shelf autopilots. Experimental results demonstrate target detection with a 69% confidence floor at an operational distance of 8.5 m. Under static conditions, Kalman filtering refined the retransmitted GPS tracking accuracy from 21 cm to 13 cm within 0.6 s. Dynamic tracking trials along a complex figure-eight trajectory demonstrated that a 100 Hz non-linear EKF—fusing 5 Hz retransmitted GPS with raw IMU variable speed—effectively neutralized indoor multipath interference. This framework achieved an exceptional 2D position RMSE of just 4.28 cm, compared to a substantial 43.20 cm error yielded by a constant-speed baseline tracking architecture. This study demonstrates that dynamic GPS retransmission provides a robust, infrastructure-light alternative to complex spatial mapping, allowing standard autonomous vehicles to navigate seamlessly within GNSS-denied environments.