搜索结果：diving

This study presents the development and experimental verification of a biomimetic manta ray robot for underwater autonomous exploration. Inspired by manta rays, the robot uses flapping motion for propulsion to minimize seabed disturbance and enhance efficiency compared to traditional screw propulsion. The robot features pectoral fins driven by servo motors and a streamlined control box to reduce fluid resistance. The control system, powered by a Raspberry Pi 3B, includes an IMU and pressure sensor for real-time monitoring and control. Experiments in a pool assessed the robot's swimming and diving capabilities. Results show stable swimming and diving motions with PD control. The robot is suitable for applications in environments like aquariums and fish nurseries, requiring minimal disturbance and efficient maneuverability. Our findings demonstrate the potential of bio-inspired robotic designs to improve ecological monitoring and underwater exploration.

Cuvier's beaked whales (Ziphius cavirostris) are the deepest diving marine mammal, consistently diving to depths exceeding 1,000m for durations longer than an hour, making them difficult animals to study. They are important to study because they are sensitive to disturbances from naval sonar. Satellite-linked telemetry devices provide up to 14-day long records of dive behavior. However, the time series of depths is discretized to coarse bins due to bandwidth limitations. We analyze telemetry data from beaked whales that were exposed to moderate levels of sonar within controlled exposure experiments (CEEs) to study behavioral responses to sound exposure. We model the data as a hidden Markov model (HMM) over the time series of discrete depth bins, introducing partially observed movement types and recent diving activity covariates to model marginal non-stationarity. Movement types provide more flexible modeling for CEEs than partially observed dive stages, which are more commonly used in dive behavior HMMs. We estimate the proposed model within a hierarchical Bayesian framework, using HMM methods to compute marginalized likelihoods and posterior predictive distributions. We assess beh

This overview paper details the findings from the Diving Deep: Forecasting Sea Surface Temperatures and Anomalies Challenge at the European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases (ECML PKDD) 2024. The challenge focused on the data-driven predictability of global sea surface temperatures (SSTs), a key factor in climate forecasting, ecosystem management, fisheries management, and climate change monitoring. The challenge involved forecasting SST anomalies (SSTAs) three months in advance using historical data and included a special task of predicting SSTAs nine months ahead for the Baltic Sea. Participants utilized various machine learning approaches to tackle the task, leveraging data from ERA5. This paper discusses the methodologies employed, the results obtained, and the lessons learned, offering insights into the future of climate-related predictive modeling.

The matching formulation makes it naturally hard for the stereo matching to handle ill-posed regions like occlusions and non-Lambertian surfaces. Fusing monocular priors has been proven helpful for ill-posed matching, but the biased monocular prior learned from small stereo datasets constrains the generalization. Recently, stereo matching has progressed by leveraging the unbiased monocular prior from the vision foundation model (VFM) to improve the generalization in ill-posed regions. We dive into the fusion process and observe three main problems limiting the fusion of the VFM monocular prior. The first problem is the misalignment between affine-invariant relative monocular depth and absolute depth of disparity. Besides, when we use the monocular feature in an iterative update structure, the over-confidence in the disparity update leads to local optima results. A direct fusion of a monocular depth map could alleviate the local optima problem, but noisy disparity results computed at the first several iterations will misguide the fusion. In this paper, we propose a binary local ordering map to guide the fusion, which converts the depth map into a binary relative format, unifying the

This study explores the impact of feathers on the hydrodynamic drag experienced by diving birds, which is critical to their foraging efficiency and survival. Employing a novel experimental approach, we analyzed the kinematics of both feathered and non-feathered projectiles during their transition from air to water using high-speed imaging and an onboard accelerometer. The drag coefficients were determined through two methods: a direct calculation from the acceleration data and a theoretical approach fitted to the observed velocity profiles. Our results indicate that feathers significantly increase the drag force during water entry, with feathered projectiles exhibiting approximately double the drag coefficient of their smooth counterparts. These findings provide new insights into the role of avian feather morphology in diving mechanics and have potential implications for the design of bio-inspired aquatic vehicles in engineering. The study also discusses the biological implications of increased drag due to feathers and suggests that factors such as body shape might play a more critical role in the diving capabilities of birds than previously understood.

In the diving competition rules, FINA specifies the code of different diving movements and its difficulty coefficient. The rule simply relies on the complexity of the action to determine the difficulty. In the formulation of the diving difficulty coefficient, the athlete's body shape has not been fully considered, so it is difficult to fully guarantee the fairness of the diving competition. Based on the above problems, this paper analyzes the rules of the FINA's 10-meter platform diving difficulty coefficient, establishes the multi-rigid-body model of the human body, obtains the relationship between the moment of inertia and the completion time of the athletes to complete each diving action and the athlete's body shape, and determines the index to measure the athlete's body shape. The Lagrange Interpolation Polynomial is used to establish the functional relationship between the body shape correction coefficient and the body shape correction index, and the body shape correction coefficient corresponding to different body type athletes is determined accordingly. Finally, a new 10-meter platform diving difficulty coefficient scheme was developed.

Semi-continuous decision variables arise naturally in many real-world applications. They are defined to take either value zero or any value within a specified range, and occur mainly to prevent small nonzero values in the solution. One particular challenge that can come with semi-continuous variables in practical models is that their upper bound may be large or even infinite. In this article, we briefly discuss these challenges, and present a new diving heuristic tailored for mixed-integer optimization problems with general semi-continuous variables. The heuristic is designed to work independently of whether the semi-continuous variables are bounded from above, and thus circumvents the specific difficulties that come with unbounded semi-continuous variables. We conduct extensive computational experiments on three different test sets, integrating the heuristic in an open-source MIP solver. The results indicate that this heuristic is a successful tool for finding high-quality solutions in negligible time. At the root node the primal gap is reduced by an average of 5 % up to 21 %, and considering the overall performance improvement, the primal integral is reduced by 2 % to 17 % on ave

Competitive diving is a well recognized aquatic sport in which a person dives from a platform or a springboard into the water. Based on the acrobatics performed during the dive, diving is classified into a finite set of action classes which are standardized by FINA. In this work, we propose an attention guided LSTM-based neural network architecture for the task of diving classification. The network takes the frames of a diving video as input and determines its class. We evaluate the performance of the proposed model on a recently introduced competitive diving dataset, Diving48. It contains over 18000 video clips which covers 48 classes of diving. The proposed model outperforms the classification accuracy of the state-of-the-art models in both 2D and 3D frameworks by 11.54% and 4.24%, respectively. We show that the network is able to localize the diver in the video frames during the dive without being trained with such a supervision.

Seeing clearly with high resolution is a foundation of Large Multimodal Models (LMMs), which has been proven to be vital for visual perception and reasoning. Existing works usually employ a straightforward resolution upscaling method, where the image consists of global and local branches, with the latter being the sliced image patches but resized to the same resolution as the former. This means that higher resolution requires more local patches, resulting in exorbitant computational expenses, and meanwhile, the dominance of local image tokens may diminish the global context. In this paper, we dive into the problems and propose a new framework as well as an elaborate optimization strategy. Specifically, we extract contextual information from the global view using a mixture of adapters, based on the observation that different adapters excel at different tasks. With regard to local patches, learnable query embeddings are introduced to reduce image tokens, the most important tokens accounting for the user question will be further selected by a similarity-based selector. Our empirical results demonstrate a `less is more' pattern, where \textit{utilizing fewer but more informative local

This work presents the first results of the Deep IFS View of Nuclei of Galaxies (DIVING$^\mathrm{3D}$) survey. We analysed the nuclear emission-line spectra of a sub-sample we call mini-DIVING$^\mathrm{3D}$, which includes all Southern galaxies with B < 11.2 and |b| > 15 degrees. We verified that $23\% \pm 4\%$ of the galaxies show nuclear emission-line properties characteristic of Low Ionization Nuclear Emission-Line Regions (LINERs). Diagnostic diagram analysis reveals an apparent dichotomy, not detected in previous studies, between objects classified as H II regions and as LINERs or Seyferts, with very few galaxies classified as transition objects. A possible explanation for this result is that at least part of the transition objects are composite systems, with a central LINER contaminated by the emission from circumnuclear H II regions. The higher spatial resolution of the DIVING$^\mathrm{3D}$ survey, in comparison with previous studies, allowed us to isolate the nuclear emission from circumnuclear contaminations, reducing the number of transition objects. We also propose an alternative scenario, in which the emission-line spectra of some transition objects are the result

Knowledge Graph(KG) grounded conversations often use large pre-trained models and usually suffer from fact hallucination. Frequently entities with no references in knowledge sources and conversation history are introduced into responses, thus hindering the flow of the conversation -- existing work attempt to overcome this issue by tweaking the training procedure or using a multi-step refining method. However, minimal effort is put into constructing an entity-level hallucination detection system, which would provide fine-grained signals that control fallacious content while generating responses. As a first step to address this issue, we dive deep to identify various modes of hallucination in KG-grounded chatbots through human feedback analysis. Secondly, we propose a series of perturbation strategies to create a synthetic dataset named FADE (FActual Dialogue Hallucination DEtection Dataset). Finally, we conduct comprehensive data analyses and create multiple baseline models for hallucination detection to compare against human-verified data and already established benchmarks.

Finding a better feasible solution in a shorter time is an integral part of solving Mixed Integer Programs. We present a post-hoc method based on Neural Diving to build heuristics more flexibly. We hypothesize that variables with higher confidence scores are more definite to be included in the optimal solution. For our hypothesis, we provide empirical evidence that confidence threshold technique produces partial solutions leading to final solutions with better primal objective values. Our method won 2nd place in the primal task on the NeurIPS 2021 ML4CO competition. Also, our method shows the best score among other learning-based methods in the competition.

Diving induces large pressures during water entry, accompanied by the creation of cavity and water splash ejected from the free water surface. To minimize impact forces, divers streamline their shape at impact. Here, we investigate the impact forces and splash evolution of diving wedges as a function of the wedge opening angle. A gradual transition from impactful to smooth entry is observed as the wedge angle decreases. After submersion, diving wedges experience significantly smaller drag forces (two-fold smaller) than immersed wedges. Our experimental findings compare favorably with existing force models upon the introduction of empirically-based corrections. We experimentally characterize the shapes of the cavity and splash created by the wedge and find that they are independent of the entry velocity at short times, but that the splash exhibits distinct variations in shape at later times. We propose a one-dimensional model of the splash that takes into account gravity, surface tension and aerodynamics forces. The model shows, in conjunction with experimental data, that the splash shape is dominated by the interplay between a destabilizing Venturi-suction force due to air rushing

We present the Deep Integral Field Spectrograph View of Nuclei of Galaxies (DIVING$^{3D}$) survey, a seeing-limited optical 3D spectroscopy study of the central regions of all 170 galaxies in the Southern hemisphere with B < 12.0 and |b| > 15 degrees. Most of the observations were taken with the Integral Field Unit of the Gemini Multi-Object Spectrograph, at the Gemini South telescope, but some are also being taken with the Southern Astrophysical Research Telescope (SOAR) Integral Field Spectrograph. The DIVING$^{3D}$ survey was designed for the study of nuclear emission-line properties, circumnuclear (within scales of hundreds of pc) emission-line properties, stellar and gas kinematics and stellar archaeology. The data have a combination of high spatial and spectral resolution not matched by previous surveys and will result in significant contributions for studies related to, for example, the statistics of low-luminosity active galactic nuclei, the ionization mechanisms in Low-Ionization Nuclear Emission-Line Regions, the nature of transition objects, among other topics.