pith. sign in

arxiv: 2606.01398 · v1 · pith:GR22Q4VWnew · submitted 2026-05-31 · 💻 cs.RO

A Sonar-Visual Dataset for Cross-Modal Underwater Robot Perception

Weitung Chen , Phil Tinn , Per Gunnar Auran , Martin Ludvigsen , Peter Halland Haro This is my paper

Pith reviewed 2026-06-28 16:45 UTC · model grok-4.3

classification 💻 cs.RO
keywords sonar-visual datasetcross-modal perceptionunderwater roboticsfish detectionpaired sensor datadata synchronizationobject detection
0
0 comments X

The pith

The SOVIS dataset supplies over 76,000 synchronized sonar-visual frame pairs that enable cross-modal fish detection with seven times the accuracy of camera-only baselines.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper sets out to remove the main obstacle to cross-modal learning in underwater robotics by releasing a large collection of accurately paired sonar and camera frames. An end-to-end cleaning and synchronization pipeline plus an annotation tool make the raw multi-sensor streams usable for training. A small labeled subset already yields a 7x gain in mean average precision at 0.10 IoU for fish detection over a monocular vision baseline. Readers should care because low visibility frequently defeats cameras alone, while sonar supplies reliable range data that could be predicted from images once enough paired examples exist. The authors present the release as the starting point for tasks such as dense sonar generation from single camera views.

Core claim

SOVIS comprises over 76,000 paired frames collected across 17 dives at six sites in the Trondheimfjord, supported by an end-to-end pipeline that cleans and synchronizes the cross-modal sensor data. An interactive annotation tool accelerates labeling of the paired data. A proof-of-concept cross-modal fish detection task using a small subset of labeled data achieves a 7x improvement in mAP@0.10 over a monocular camera baseline, positioning the dataset as the first step toward dense sonar prediction from monocular images.

What carries the argument

The end-to-end pipeline that cleans and synchronizes cross-modal sensor streams to produce accurately paired sonar-visual frames for supervised learning.

If this is right

  • Models trained on the pairs can combine visual semantics with acoustic range to detect fish more reliably than vision alone.
  • The dataset directly supports experiments that attempt to predict full sonar outputs from single monocular images.
  • The synchronization pipeline supplies a reusable method for constructing additional multi-modal underwater collections.
  • Larger volumes of paired examples lower the barrier to developing cross-modal algorithms for robot perception.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The same pairs could be used to train models that correct visibility loss in one modality by reference to the other.
  • Performance improvements observed on fish may extend to detection of other objects such as structures or debris once more labels are added.
  • Real-robot deployment would reveal whether the learned mappings remain stable under motion and lighting changes absent from the static dives.
  • The six sites may or may not represent the full range of turbidity and bottom types encountered in open-ocean operations.

Load-bearing premise

The pipeline produces correctly paired frames without misalignment errors or biases that would distort what models learn from the data.

What would settle it

Retraining the fish detector after independently re-synchronizing the original raw sensor logs and finding that the reported sevenfold mAP gain disappears.

Figures

Figures reproduced from arXiv: 2606.01398 by Martin Ludvigsen, Per Gunnar Auran, Peter Halland Haro, Phil Tinn, Weitung Chen.

Figure 1
Figure 1. Figure 1: SOVIS overview. A Blueye X3 ROV equipped with a monocular camera (green FoV) and an Oculus M750d multibeam sonar (blue FoV) collects synchronized underwater data. (a) Camera frame showing fish right above the seafloor. (b) Corresponding sonar return of the same scene. platforms, dive operations, and manual synchronization of sensors with fundamentally different coordinate systems. Furthermore, simulated da… view at source ↗
Figure 2
Figure 2. Figure 2: Data collection platform. The Blueye X3 ROV with a tether reel. The ROV carries a monocular camera, an Oculus M750d multibeam sonar, and onboard temperature and pressure sensor. III. SONAR-VISUAL DATASET SOVIS is a large-scale, synchronized sonar-visual dataset consisting of time-aligned monocular camera images and multibeam sonar returns, paired with environmental mea￾surements (water temperature and pres… view at source ↗
Figure 3
Figure 3. Figure 3: Deployment sites. Satellite view of six data collection sites (A–F) in the Trondheimfjord, Norway. Site labels indicate the number of dives conducted at each location. TABLE II Summary statistics of the SOVIS dataset. Attribute Value Total dives 17 Deployment sites 6 (Trondheimfjord) Total recording hours 2.1 h Synchronized pairs ∼76,600 Camera resolution 1920×1080 Sonar operating freq. 750 kHz Water temp.… view at source ↗
Figure 4
Figure 4. Figure 4: An example sonar-visual pair. A monocular camera frame (top) and its temporally aligned Oculus M750d sonar return (bottom), captured simultaneously during a dive. A. Interface Design The tool presents synchronized camera and sonar views side-by-side ( [PITH_FULL_IMAGE:figures/full_fig_p004_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: Cross-modal annotation tool. The interface displays the camera frame (left) with fish bounding boxes in cyan and the sonar fan visualization (top right) with angular bands highlighting detected regions. The bottom strip shows ±2 neighboring sonar frames for temporal context. When an annotator draws a bounding box on the camera image, the tool automatically projects it to the corresponding angular extent in… view at source ↗
Figure 6
Figure 6. Figure 6: SOVISFishNet architecture. An EfficientNet-B4 encoder extracts multi-scale features, which are fused by an FPN neck into a unified feature map. A 2D Gaussian spatial attention mask, generated from the camera fish bounding box, gates the feature map. After global average pooling (GAP), the visual features are concatenated with the normalized bounding box coordinates and decoded by an MLP head into four sona… view at source ↗
Figure 7
Figure 7. Figure 7: Qualitative cross-modal detection results. Each panel shows the camera frame (left) with detected fish (green box) and the sonar fan view (right) with ground-truth (green) and predicted (red) bounding boxes. The model correctly localizes the acoustic signature of the visually detected fish, achieving IoU of 0.60 (top) and 0.50 (bottom). mapping, but it cannot resolve range and simply defaults to the traini… view at source ↗
read the original abstract

Underwater robots typically use both cameras and sonar for perception to leverage the rich semantic details of vision and the robust range measurements of acoustics. However, learning to map between these modalities via cross-modal prediction remains underexplored due to limited sonar-visual paired datasets. We present SOVIS, a sonar-visual dataset for cross-modal underwater perception. SOVIS comprises over 76,000 paired frames collected across 17 dives at six sites in the Trondheimfjord, supported by an end-to-end pipeline that cleans and synchronizes the cross-modal sensor data. We also introduce an interactive annotation tool designed to accelerate the labeling process for this paired data. Finally, we demonstrate a proof-of-concept cross-modal fish detection task using a small subset of labeled data, achieving a 7x improvement in mAP@0.10 over a monocular camera baseline. SOVIS serves as the first step toward advancing cross-modal underwater perception research, enabling research directions such as dense sonar prediction from monocular images.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 1 minor

Summary. The paper presents SOVIS, a sonar-visual dataset comprising over 76,000 paired frames collected across 17 dives at six sites in the Trondheimfjord. It describes an end-to-end pipeline for cleaning and synchronizing the cross-modal sensor data, introduces an interactive annotation tool, and demonstrates a proof-of-concept cross-modal fish detection task on a small labeled subset that achieves a 7x improvement in mAP@0.10 over a monocular camera baseline.

Significance. If the pairing quality holds, the dataset and annotation tool would provide a valuable empirical resource for cross-modal underwater perception, addressing the scarcity of large-scale paired sonar-visual data and enabling downstream tasks such as dense prediction and fish detection. The manuscript's strength lies in its scale of data collection and the provision of a practical pipeline and tool.

major comments (2)
  1. [§3] §3: The end-to-end pipeline for timestamp matching, cleaning heuristics, and synchronization is described in detail, but the manuscript reports no quantitative validation metrics such as mean/max timestamp offset (in ms), fraction of pairs rejected due to misalignment, or statistics from manual spot-checks. This is load-bearing for the central claim that the >76k frames supply reliably paired data usable for cross-modal learning.
  2. [Proof-of-concept results] Proof-of-concept results: The reported 7x mAP@0.10 improvement is evaluated on a small labeled subset, yet the text provides no details on baseline implementation, train/test splits, error bars, or confirmation that synchronization errors were not a confounding factor in the cross-modal signal. This leaves the empirical demonstration weakly supported.
minor comments (1)
  1. The abstract states the 7x improvement but does not indicate the size of the labeled subset used, which would aid reader assessment of the result's scope.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the constructive feedback. We address each major comment below and commit to revisions that strengthen the manuscript's claims regarding data quality and the proof-of-concept evaluation.

read point-by-point responses

  1. Referee: [§3] §3: The end-to-end pipeline for timestamp matching, cleaning heuristics, and synchronization is described in detail, but the manuscript reports no quantitative validation metrics such as mean/max timestamp offset (in ms), fraction of pairs rejected due to misalignment, or statistics from manual spot-checks. This is load-bearing for the central claim that the >76k frames supply reliably paired data usable for cross-modal learning.

    Authors: We agree that quantitative validation metrics are necessary to support the reliability of the paired data. In the revised manuscript we will augment Section 3 with the requested statistics: the distribution (mean, max, std) of timestamp offsets across all pairs, the fraction of candidate pairs rejected by each cleaning heuristic, and quantitative results from manual spot-checks performed on a random sample of 500 pairs. These additions will directly address the load-bearing concern. revision: yes

  2. Referee: [Proof-of-concept results] Proof-of-concept results: The reported 7x mAP@0.10 improvement is evaluated on a small labeled subset, yet the text provides no details on baseline implementation, train/test splits, error bars, or confirmation that synchronization errors were not a confounding factor in the cross-modal signal. This leaves the empirical demonstration weakly supported.

    Authors: We acknowledge the need for greater transparency in the proof-of-concept. The revised text will specify the exact baseline architecture and training protocol, the train/validation/test split ratios and sizes for the labeled subset, standard deviations or error bars across repeated runs, and a targeted analysis (e.g., ablation on deliberately mis-synchronized pairs) confirming that the observed 7x gain is not an artifact of residual synchronization error. These details will be added to the relevant experimental section. revision: yes

Circularity Check

0 steps flagged

No circularity: empirical dataset paper with no derivation chain

full rationale

The manuscript describes collection of a real-world sonar-visual dataset across 17 dives, an end-to-end cleaning/synchronization pipeline, an annotation tool, and a small proof-of-concept detection demo. No equations, fitted parameters, uniqueness theorems, or self-citations are invoked to derive any result from prior outputs of the same work. The 7× mAP improvement is reported as an empirical observation on the collected data rather than a prediction forced by construction. The synchronization pipeline is described procedurally but is not presented as a mathematical derivation that reduces to its own inputs. This is a standard data-release contribution whose central claims rest on external measurement rather than internal self-reference.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

Based solely on the abstract, the work involves no mathematical model, fitted parameters, axioms, or invented entities; it is a data collection and tooling contribution.

pith-pipeline@v0.9.1-grok · 5715 in / 1099 out tokens · 36644 ms · 2026-06-28T16:45:57.911666+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

123 extracted references · 2 canonical work pages · 1 internal anchor

  1. [1]

    Science Robotics , volume=

    Acrobatics at the insect scale: A durable, precise, and agile micro--aerial robot , author=. Science Robotics , volume=. 2025 , publisher=

    2025

  2. [2]

    IEEE Transactions on Robotics , volume=

    A comparative study of nonlinear mpc and differential-flatness-based control for quadrotor agile flight , author=. IEEE Transactions on Robotics , volume=. 2022 , publisher=

    2022

  3. [3]

    IEEE Robotics and Automation Letters , volume=

    Passive wall tracking for a rotorcraft with tilted and ducted propellers using proximity effects , author=. IEEE Robotics and Automation Letters , volume=. 2022 , publisher=

    2022

  4. [4]

    Science , volume=

    The aerodynamics of free-flight maneuvers in Drosophila , author=. Science , volume=. 2003 , publisher=

    2003

  5. [5]

    Proceedings of the National Academy of Sciences , volume=

    Discovering the flight autostabilizer of fruit flies by inducing aerial stumbles , author=. Proceedings of the National Academy of Sciences , volume=. 2010 , publisher=

    2010

  6. [6]

    Frontiers in neuroscience , volume=

    Comparison of visually guided flight in insects and birds , author=. Frontiers in neuroscience , volume=. 2018 , publisher=

    2018

  7. [7]

    Science , volume=

    A tailless aerial robotic flapper reveals that flies use torque coupling in rapid banked turns , author=. Science , volume=. 2018 , publisher=

    2018

  8. [8]

    IEEE Robotics and Automation Letters , volume=

    Flying with damaged wings: The effect on flight capacity and bio-inspired coping strategies of a flapping wing robot , author=. IEEE Robotics and Automation Letters , volume=. 2021 , publisher=

    2021

  9. [9]

    2015 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) , pages=

    Rotating the heading angle of underactuated flapping-wing flyers by wriggle-steering , author=. 2015 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) , pages=. 2015 , organization=

    2015

  10. [10]

    Science , volume=

    Aggressive mimicry in Photuris: firefly femmes fatales , author=. Science , volume=. 1965 , publisher=

    1965

  11. [11]

    The Biological Bulletin , volume=

    Courting behavior in a synchronously flashing, aggregative firefly, Pteroptyx tener , author=. The Biological Bulletin , volume=. 1980 , publisher=

    1980

  12. [12]

    The American Naturalist , volume=

    Energy and predation costs of firefly courtship signals , author=. The American Naturalist , volume=. 2007 , publisher=

    2007

  13. [13]

    Science , volume=

    Controlled flight of a biologically inspired, insect-scale robot , author=. Science , volume=. 2013 , publisher=

    2013

  14. [14]

    IEEE Robotics and Automation Letters , volume=

    Toward controlled flight of the ionocraft: a flying microrobot using electrohydrodynamic thrust with onboard sensing and no moving parts , author=. IEEE Robotics and Automation Letters , volume=. 2018 , publisher=

    2018

  15. [15]

    IEEE Robotics and Automation Letters , volume=

    Bee+: A 95-mg four-winged insect-scale flying robot driven by twinned unimorph actuators , author=. IEEE Robotics and Automation Letters , volume=. 2019 , publisher=

    2019

  16. [16]

    IEEE Transactions on Robotics , volume=

    Liftoff of an electromagnetically driven insect-inspired flapping-wing robot , author=. IEEE Transactions on Robotics , volume=. 2016 , publisher=

    2016

  17. [17]

    Science Robotics , volume=

    A biologically inspired, flapping-wing, hybrid aerial-aquatic microrobot , author=. Science Robotics , volume=. 2017 , publisher=

    2017

  18. [18]

    IEEE Transactions on Robotics , volume=

    RoboFly: An insect-sized robot with simplified fabrication that is capable of flight, ground, and water surface locomotion , author=. IEEE Transactions on Robotics , volume=. 2021 , publisher=

    2021

  19. [19]

    Science , volume=

    Perching and takeoff of a robotic insect on overhangs using switchable electrostatic adhesion , author=. Science , volume=. 2016 , publisher=

    2016

  20. [20]

    2014 IEEE international conference on robotics and automation (ICRA) , pages=

    Pitch and yaw control of a robotic insect using an onboard magnetometer , author=. 2014 IEEE international conference on robotics and automation (ICRA) , pages=. 2014 , organization=

    2014

  21. [21]

    Nature , volume=

    Untethered flight of an insect-sized flapping-wing microscale aerial vehicle , author=. Nature , volume=. 2019 , publisher=

    2019

  22. [22]

    2018 IEEE International Conference on Robotics and Automation (ICRA) , pages=

    Liftoff of a 190 mg laser-powered aerial vehicle: The lightest wireless robot to fly , author=. 2018 IEEE International Conference on Robotics and Automation (ICRA) , pages=. 2018 , organization=

    2018

  23. [23]

    Advanced Materials , pages=

    A High-Lift Micro-Aerial-Robot Powered by Low-Voltage and Long-Endurance Dielectric Elastomer Actuators , author=. Advanced Materials , pages=. 2022 , publisher=

    2022

  24. [24]

    Nature , volume=

    Controlled flight of a microrobot powered by soft artificial muscles , author=. Nature , volume=. 2019 , publisher=

    2019

  25. [25]

    IEEE Transactions on Robotics , volume=

    Collision resilient insect-scale soft-actuated aerial robots with high agility , author=. IEEE Transactions on Robotics , volume=. 2021 , publisher=

    2021

  26. [26]

    Advanced Materials , volume=

    Highly stretchable and self-deformable alternating current electroluminescent devices , author=. Advanced Materials , volume=. 2015 , publisher=

    2015

  27. [27]

    Advanced Materials , volume=

    Bright stretchable alternating current electroluminescent displays based on high permittivity composites , author=. Advanced Materials , volume=. 2016 , publisher=

    2016

  28. [28]

    Journal of Materials Chemistry C , volume=

    Highly bright and stable electroluminescent devices with extraordinary stretchability and ultraconformability , author=. Journal of Materials Chemistry C , volume=. 2019 , publisher=

    2019

  29. [29]

    science , volume=

    Highly stretchable electroluminescent skin for optical signaling and tactile sensing , author=. science , volume=. 2016 , publisher=

    2016

  30. [30]

    Soft Matter , volume=

    Electroluminescent soft elastomer actuators with adjustable luminance and strain , author=. Soft Matter , volume=. 2019 , publisher=

    2019

  31. [31]

    Advanced Materials , volume=

    Multilayer dielectric elastomers for fast, programmable actuation without prestretch , author=. Advanced Materials , volume=. 2016 , publisher=

    2016

  32. [32]

    2003 , publisher=

    Multiple view geometry in computer vision , author=. 2003 , publisher=

    2003

  33. [33]

    Journal of Applied Physics , volume=

    A high-performance dielectric elastomer consisting of bio-based polyester elastomer and titanium dioxide powder , author=. Journal of Applied Physics , volume=. 2013 , publisher=

    2013

  34. [34]

    2014 , publisher=

    Using a MEMS gyroscope to stabilize the attitude of a fly-sized hovering robot , author=. 2014 , publisher=

    2014

  35. [35]

    Incremental Nonlinear Fault-Tolerant Control of a Quadrotor With Complete Loss of Two Opposing Rotors , year=

    Sun, Sihao and Wang, Xuerui and Chu, Qiping and Visser, Coen de , journal=. Incremental Nonlinear Fault-Tolerant Control of a Quadrotor With Complete Loss of Two Opposing Rotors , year=

  36. [36]

    High-Speed Flight of Quadrotor Despite Loss of Single Rotor , year=

    Sun, Sihao and Sijbers, Leon and Wang, Xuerui and de Visser, Coen , journal=. High-Speed Flight of Quadrotor Despite Loss of Single Rotor , year=

  37. [37]

    Single-loop control and trajectory following of a flapping-wing microrobot , year=

    Chirarattananon, Pakpong and Ma, Kevin Y and Wood, Robert J , booktitle=. Single-loop control and trajectory following of a flapping-wing microrobot , year=

  38. [38]

    IEEE Robotics and Automation Letters , volume=

    Yaw control of a hovering flapping-wing aerial vehicle with a passive wing hinge , author=. IEEE Robotics and Automation Letters , volume=. 2021 , publisher=

    2021

  39. [39]

    Interface focus , volume=

    Dynamics and flight control of a flapping-wing robotic insect in the presence of wind gusts , author=. Interface focus , volume=. 2017 , publisher=

    2017

  40. [40]

    In 2014 IEEE , author=

    Principles of microscale flexure hinge design for enhanced endurance. In 2014 IEEE , author=. RSJ International Conference on Intelligent Robots and Systems (IROS 2014) , pages=

    2014

  41. [41]

    IEEE Robotics and Automation Letters , volume=

    Four wings: An insect-sized aerial robot with steering ability and payload capacity for autonomy , author=. IEEE Robotics and Automation Letters , volume=. 2019 , publisher=

    2019

  42. [42]

    Journal of Fluid Mechanics , volume=

    Experimental and computational studies of the aerodynamic performance of a flapping and passively rotating insect wing , author=. Journal of Fluid Mechanics , volume=. 2016 , publisher=

    2016

  43. [43]

    Journal of Fluid Mechanics , volume=

    Passive wing pitch reversal in insect flight , author=. Journal of Fluid Mechanics , volume=. 2007 , publisher=

    2007

  44. [44]

    Journal of fluid mechanics , volume=

    Influence of wing kinematics on aerodynamic performance in hovering insect flight , author=. Journal of fluid mechanics , volume=. 2008 , publisher=

    2008

  45. [45]

    IEEE Robotics and Automation Letters , volume=

    Modeling and Control of Flapping-Wing Micro-Aerial Vehicles With Harmonic Sinusoids , author=. IEEE Robotics and Automation Letters , volume=. 2021 , publisher=

    2021

  46. [46]

    and Helbling, E

    Steinmeyer, Rebecca and Hyun, Nak-seung P. and Helbling, E. Farrell and Wood, Robert J. , booktitle=. Yaw Torque Authority for a Flapping-Wing Micro-Aerial Vehicle , year=

  47. [47]

    The International Journal of Robotics Research , pages=

    An efficient, modular controller for flapping flight composing model-based and model-free components , author=. The International Journal of Robotics Research , pages=. 2021 , publisher=

    2021

  48. [48]

    Bioinspiration & biomimetics , volume=

    Adaptive control of a millimeter-scale flapping-wing robot , author=. Bioinspiration & biomimetics , volume=. 2014 , publisher=

    2014

  49. [49]

    and Wood, Robert J

    Chirarattananon, Pakpong and Ma, Kevin Y. and Wood, Robert J. , booktitle=. Fly on the wall , year=

  50. [50]

    The International Journal of Robotics Research , volume=

    Perching with a robotic insect using adaptive tracking control and iterative learning control , author=. The International Journal of Robotics Research , volume=. 2016 , publisher=

    2016

  51. [51]

    , journal=

    Wood, Robert J. , journal=. The First Takeoff of a Biologically Inspired At-Scale Robotic Insect , year=

  52. [52]

    2011 IEEE international conference on robotics and automation , pages=

    Minimum snap trajectory generation and control for quadrotors , author=. 2011 IEEE international conference on robotics and automation , pages=. 2011 , organization=

    2011

  53. [53]

    The International Journal of Robotics Research , volume=

    Trajectory generation and control for precise aggressive maneuvers with quadrotors , author=. The International Journal of Robotics Research , volume=. 2012 , publisher=

    2012

  54. [54]

    AIAA Guidance, Navigation and Control Conference and Exhibit , pages=

    Experiments in fixed-wing UAV perching , author=. AIAA Guidance, Navigation and Control Conference and Exhibit , pages=

  55. [55]

    IEEE Robotics and Automation Letters , volume=

    FireFly: An Insect-Scale Aerial Robot Powered by Electroluminescent Soft Artificial Muscles , author=. IEEE Robotics and Automation Letters , volume=. 2022 , publisher=

    2022

  56. [56]

    2015 , school=

    Design of hybrid passive and active mechanisms for control of insect-scale flapping-wing robots , author=. 2015 , school=

    2015

  57. [57]

    Journal of mathematical Biology , volume=

    Mathematical model of honeycomb construction , author=. Journal of mathematical Biology , volume=. 1986 , publisher=

    1986

  58. [58]

    Oecologia , volume=

    Bumble bee behavior and selection on flower size in the sky pilot, Polemonium viscosum , author=. Oecologia , volume=. 1987 , publisher=

    1987

  59. [59]

    Proceedings of the National Academy of Sciences , volume=

    Mosquitoes survive raindrop collisions by virtue of their low mass , author=. Proceedings of the National Academy of Sciences , volume=. 2012 , publisher=

    2012

  60. [60]

    Science advances , volume=

    Flies land upside down on a ceiling using rapid visually mediated rotational maneuvers , author=. Science advances , volume=. 2019 , publisher=

    2019

  61. [61]

    Journal of Micromechanics and Microengineering , volume=

    Pop-up book MEMS , author=. Journal of Micromechanics and Microengineering , volume=. 2011 , publisher=

    2011

  62. [62]

    2013 16th International Conference on Advanced Robotics (ICAR) , pages=

    Model-free control of a flapping-wing flying microrobot , author=. 2013 16th International Conference on Advanced Robotics (ICAR) , pages=. 2013 , organization=

    2013

  63. [63]

    Science Robotics , volume=

    A gyroscope-free visual-inertial flight control and wind sensing system for 10-mg robots , author=. Science Robotics , volume=. 2022 , publisher=

    2022

  64. [64]

    Microrobot design using fiber reinforced composites , author=

  65. [65]

    Journal of Micromechanics and Microengineering , volume=

    Monolithic fabrication of millimeter-scale machines , author=. Journal of Micromechanics and Microengineering , volume=. 2012 , publisher=

    2012

  66. [66]

    Smart Materials and Structures , volume=

    Design and manufacturing rules for maximizing the performance of polycrystalline piezoelectric bending actuators , author=. Smart Materials and Structures , volume=. 2015 , publisher=

    2015

  67. [67]

    Micro-and Nanotechnology Sensors, Systems, and Applications VII , volume=

    PopupCAD: a tool for automated design, fabrication, and analysis of laminate devices , author=. Micro-and Nanotechnology Sensors, Systems, and Applications VII , volume=. 2015 , organization=

    2015

  68. [68]

    International Design Engineering Technical Conferences and Computers and Information in Engineering Conference , volume=

    An integrated design and simulation environment for rapid prototyping of laminate robotic mechanisms , author=. International Design Engineering Technical Conferences and Computers and Information in Engineering Conference , volume=. 2018 , organization=

    2018

  69. [69]

    2016 IEEE International Conference on Robotics and Automation (ICRA) , pages=

    Non-linear resonance modeling and system design improvements for underactuated flapping-wing vehicles , author=. 2016 IEEE International Conference on Robotics and Automation (ICRA) , pages=. 2016 , organization=

    2016

  70. [70]

    2011 IEEE/RSJ International Conference on Intelligent Robots and Systems , pages=

    System identification and linear time-invariant modeling of an insect-sized flapping-wing micro air vehicle , author=. 2011 IEEE/RSJ International Conference on Intelligent Robots and Systems , pages=. 2011 , organization=

    2011

  71. [71]

    2016 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) , pages=

    Influence of wing morphological and inertial parameters on flapping flight performance , author=. 2016 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) , pages=. 2016 , organization=

    2016

  72. [72]

    2018 IEEE International Conference on Real-time Computing and Robotics (RCAR) , pages=

    An efficient method for the design and fabrication of 2D laminate robotic structures , author=. 2018 IEEE International Conference on Real-time Computing and Robotics (RCAR) , pages=. 2018 , organization=

    2018

  73. [73]

    IEEE Robotics and Automation Letters , volume=

    Scalable cooperative transport of cable-suspended loads with UAVs using distributed trajectory optimization , author=. IEEE Robotics and Automation Letters , volume=. 2020 , publisher=

    2020

  74. [74]

    2008 IEEE international conference on robotics and automation , pages=

    Reciprocal velocity obstacles for real-time multi-agent navigation , author=. 2008 IEEE international conference on robotics and automation , pages=. 2008 , organization=

    2008

  75. [75]

    IEEE Transactions on Robotics , volume=

    Liftoff of a motor-driven, flapping-wing microaerial vehicle capable of resonance , author=. IEEE Transactions on Robotics , volume=. 2013 , publisher=

    2013

  76. [76]

    Science , volume=

    Programmable self-assembly in a thousand-robot swarm , author=. Science , volume=. 2014 , publisher=

    2014

  77. [77]

    IEEE Transactions on Robotics , volume=

    A survey on aerial swarm robotics , author=. IEEE Transactions on Robotics , volume=. 2018 , publisher=

    2018

  78. [78]

    Frontiers in Robotics and AI , volume=

    A survey on swarming with micro air vehicles: Fundamental challenges and constraints , author=. Frontiers in Robotics and AI , volume=. 2020 , publisher=

    2020

  79. [79]

    Nature Machine Intelligence , volume=

    Predictive control of aerial swarms in cluttered environments , author=. Nature Machine Intelligence , volume=. 2021 , publisher=

    2021

  80. [80]

    Science Robotics , volume=

    Minimal navigation solution for a swarm of tiny flying robots to explore an unknown environment , author=. Science Robotics , volume=. 2019 , publisher=

    2019

Showing first 80 references.