Girdhar Yogesh

No Thumbnail Available
Last Name
First Name

Search Results

Now showing 1 - 6 of 6
  • Article
    A crab swarm at an ecological hotspot : patchiness and population density from AUV observations at a coastal, tropical seamount
    (PeerJ, 2016-04-12) Pineda, Jesus ; Cho, Walter W. ; Starczak, Victoria R. ; Govindarajan, Annette F. ; Guzman, Hector M. ; Girdhar, Yogesh ; Holleman, Rusty C. ; Churchill, James H. ; Singh, Hanumant ; Ralston, David K.
    A research cruise to Hannibal Bank, a seamount and an ecological hotspot in the coastal eastern tropical Pacific Ocean off Panama, explored the zonation, biodiversity, and the ecological processes that contribute to the seamount’s elevated biomass. Here we describe the spatial structure of a benthic anomuran red crab population, using submarine video and autonomous underwater vehicle (AUV) photographs. High density aggregations and a swarm of red crabs were associated with a dense turbid layer 4–10 m above the bottom. The high density aggregations were constrained to 355–385 m water depth over the Northwest flank of the seamount, although the crabs also occurred at lower densities in shallower waters (∼280 m) and in another location of the seamount. The crab aggregations occurred in hypoxic water, with oxygen levels of 0.04 ml/l. Barcoding of Hannibal red crabs, and pelagic red crabs sampled in a mass stranding event in 2015 at a beach in San Diego, California, USA, revealed that the Panamanian and the Californian crabs are likely the same species, Pleuroncodes planipes, and these findings represent an extension of the southern endrange of this species. Measurements along a 1.6 km transect revealed three high density aggregations, with the highest density up to 78 crabs/m2, and that the crabs were patchily distributed. Crab density peaked in the middle of the patch, a density structure similar to that of swarming insects.
  • Article
    Detection of unanticipated faults for autonomous underwater vehicles using online topic models
    (John Wiley & Sons, 2017-12-26) Raanan, Ben-Yair ; Bellingham, James G. ; Zhang, Yanwu ; Kemp, Mathieu ; Kieft, Brian ; Singh, Hanumant ; Girdhar, Yogesh
    For robots to succeed in complex missions, they must be reliable in the face of subsystem failures and environmental challenges. In this paper, we focus on autonomous underwater vehicle (AUV) autonomy as it pertains to self‐perception and health monitoring, and we argue that automatic classification of state‐sensor data represents an important enabling capability. We apply an online Bayesian nonparametric topic modeling technique to AUV sensor data in order to automatically characterize its performance patterns, then demonstrate how in combination with operator‐supplied semantic labels these patterns can be used for fault detection and diagnosis by means of a nearest‐neighbor classifier. The method is evaluated using data collected by the Monterey Bay Aquarium Research Institute's Tethys long‐range AUV in three separate field deployments. Our results show that the proposed method is able to accurately identify and characterize patterns that correspond to various states of the AUV, and classify faults at a high rate of correct detection with a very low false detection rate.
  • Preprint
    Modeling curiosity in a mobile robot for long-term autonomous exploration and monitoring
    ( 2015-09) Girdhar, Yogesh ; Dudek, Gregory
    This paper presents a novel approach to modeling curiosity in a mobile robot, which is useful for monitoring and adaptive data collection tasks, especially in the context of long term autonomous missions where pre- programmed missions are likely to have limited utility. We use a realtime topic modeling technique to build a semantic perception model of the environment, using which, we plan a path through the locations in the world with high se- mantic information content. The life-long learning behavior of the proposed perception model makes it suitable for long-term exploration missions. We val- idate the approach using simulated exploration experiments using aerial and underwater data, and demonstrate an implementation on the Aqua underwa- ter robot in a variety of scenarios. We nd that the proposed exploration paths that are biased towards locations with high topic perplexity, produce better terrain models with high discriminative power. Moreover, we show that the proposed algorithm implemented on Aqua robot is able to do tasks such as coral reef inspection, diver following, and sea oor exploration, without any prior training or preparation.
  • Article
    Assessment of attraction and avoidance behaviors of fish in response to the proximity of transiting underwater vehicles
    (National Marine Fisheries Service, 2021-11-01) Campbell, Matthew D ; Huddleston, Ariane ; Somerton, David ; Clarke, M. Elizabeth ; Wakefield, Waldo ; Murawski, Steve ; Taylor, Chris ; Singh, Hanumant ; Girdhar, Yogesh ; Yoklavich, Mary
    Underwater vehicles have many advantages for sampling fish; however, estimates can be biased by behavioral responses to sampling gear. To evaluate avoidance and attraction bias we assessed changes in fish abundance relative to a variety of sampling vehicles during transit through a test bed. Fish species were classified into five attraction and avoidance categories according to the behavioral responses exhibited. We observed that the rigor of behavioral responses varied by vehicle, vehicle range and altitude, transect number, and habitat complexity. The effect of each variable is dependent on behavioral guild, but vehicle range was the most consistent predictor of changes in abundance regardless of vehicle. Vehicles that surveyed the environment at higher relative altitudes off the seafloor and at slower speeds elicited weaker behavioral responses regardless of whether those reactions were attraction or avoidance. The test-bed approach allowed assessment of responses that cannot be observed from the perspective of a sampling vehicle but was restricted by the number of species-specific interactions observed. Despite success in estimating behavioral responses, calibrating the effect against known densities of fish was not possible. However, the method used is a robust way for future investigations to quantify species-specific responses for gear calibration and to provide information that aids in the calculation of fish abundance.
  • Article
    Toward a new era of coral reef monitoring
    (American Chemical Society, 2023-03-17) Apprill, Amy ; Girdhar, Yogesh ; Mooney, T. Aran ; Hansel, Colleen M. ; Long, Matthew H. ; Liu, Yaqin ; Zhang, W. Gordon ; Kapit, Jason ; Hughen, Konrad ; Coogan, Jeff ; Greene, Austin
    Coral reefs host some of the highest concentrations of biodiversity and economic value in the oceans, yet these ecosystems are under threat due to climate change and other human impacts. Reef monitoring is routinely used to help prioritize reefs for conservation and evaluate the success of intervention efforts. Reef status and health are most frequently characterized using diver-based surveys, but the inherent limitations of these methods mean there is a growing need for advanced, standardized, and automated reef techniques that capture the complex nature of the ecosystem. Here we draw on experiences from our own interdisciplinary research programs to describe advances in in situ diver-based and autonomous reef monitoring. We present our vision for integrating interdisciplinary measurements for select “case-study” reefs worldwide and for learning patterns within the biological, physical, and chemical reef components and their interactions. Ultimately, these efforts could support the development of a scalable and standardized suite of sensors that capture and relay key data to assist in categorizing reef health. This framework has the potential to provide stakeholders with the information necessary to assess reef health during an unprecedented time of reef change as well as restoration and intervention activities.
  • Article
    Semi-supervised visual tracking of marine animals using autonomous underwater vehicles
    (Springer, 2023-03-01) Cai, Levi ; McGuire, Nathan E. ; Hanlon, Roger ; T Aran Mooney ; Girdhar, Yogesh
    In-situ visual observations of marine organisms is crucial to developing behavioural understandings and their relations to their surrounding ecosystem. Typically, these observations are collected via divers, tags, and remotely-operated or human-piloted vehicles. Recently, however, autonomous underwater vehicles equipped with cameras and embedded computers with GPU capabilities are being developed for a variety of applications, and in particular, can be used to supplement these existing data collection mechanisms where human operation or tags are more difficult. Existing approaches have focused on using fully-supervised tracking methods, but labelled data for many underwater species are severely lacking. Semi-supervised trackers may offer alternative tracking solutions because they require less data than fully-supervised counterparts. However, because there are not existing realistic underwater tracking datasets, the performance of semi-supervised tracking algorithms in the marine domain is not well understood. To better evaluate their performance and utility, in this paper we provide (1) a novel dataset specific to marine animals located at, (2) an evaluation of state-of-the-art semi-supervised algorithms in the context of underwater animal tracking, and (3) an evaluation of real-world performance through demonstrations using a semi-supervised algorithm on-board an autonomous underwater vehicle to track marine animals in the wild.