More Like this

1. Comparison of CNN and LSTM Networks on Human Intention Prediction in Physical Human-Robot Interactions

   Zadeh, Khosro G; Zendehdel, Niloofar; Holmes, George L; Bonnett, Keyri M; Costa, Amy; Burns, Devin; Leu, Ming C; Song, Yun Seong (August 2025, The Fool)

   Advancements in robotics and AI have increased the demand for interactive robots in healthcare and assistive applications. However, ensuring safe and effective physical human-robot interactions (pHRIs) remains challenging due to the complexities of human motor communication and intent recognition. Traditional physics-based models struggle to capture the dynamic nature of human force interactions, limiting robotic adaptability. To address these limitations, neural networks (NNs) have been explored for force-movement intention prediction. While multi-layer perceptron (MLP) networks show potential, they struggle with temporal dependencies and generalization. Long Short-Term Memory (LSTM) networks effectively model sequential dependencies, while Convolutional Neural Networks (CNNs) enhance spatial feature extraction from human force data. Building on these strengths, this study introduces a hybrid LSTM-CNN framework to improve force-movement intention prediction, increasing accuracy from 69% to 86% through effective denoising and advanced architectures. The combined CNN-LSTM network proved particularly effective in handling individualized force-velocity relationships and presents a generalizable model paving the way for more adaptive strategies in robot guidance. These findings highlight the importance of integrating spatial and temporal modeling to enhance robot precision, responsiveness, and human-robot collaboration. Index Terms —- Physical Human-Robot Interaction, Intention Detection, Machine Learning, Long-Short Term Memory (LSTM) 

   more » « less
2. Cross-Hemispheric Genetic Diversity and Spatial Genetic Structure of Callinectes sapidus Reovirus 1 (CsRV1)

   https://doi.org/10.3390/v15020563  

   Zhao, Mingli; Plough, Louis V.; Behringer, Donald C.; Bojko, Jamie; Kough, Andrew S.; Alper, Nathaniel W.; Xu, Lan; Schott, Eric J. (February 2023, Viruses)

   The movement of viruses in aquatic systems is rarely studied over large geographic scales. Oceanic currents, host migration, latitude-based variation in climate, and resulting changes in host life history are all potential drivers of virus connectivity, adaptation, and genetic structure. To expand our understanding of the genetic diversity of Callinectes sapidus reovirus 1 (CsRV1) across a broad spatial and host life history range of its blue crab host (Callinectes sapidus), we obtained 22 complete and 96 partial genomic sequences for CsRV1 strains from the US Atlantic coast, Gulf of Mexico, Caribbean Sea, and the Atlantic coast of South America. Phylogenetic analyses of CsRV1 genomes revealed that virus genotypes were divided into four major genogroups consistent with their host geographic origins. However, some CsRV1 sequences from the US mid-Atlantic shared high genetic similarity with the Gulf of Mexico genotypes, suggesting potential human-mediated movement of CsRV1 between the US mid-Atlantic and Gulf coasts. This study advances our understanding of how climate, coastal geography, host life history, and human activity drive patterns of genetic structure and diversity of viruses in marine animals and contributes to the capacity to infer broadscale host population connectivity in marine ecosystems from virus population genetic data. 

   more » « less
3. Social-ecological models with social hierarchy and spatial structure applied to small-scale fisheries

   https://doi.org/10.1007/s12080-024-00594-4  

   Wulfing, Sophie; White, Easton R (December 2024, Theoretical Ecology)

   Socio-ecological models combine ecological systems with human social dynamics in order to better understand human interactions with the environment. To model human behavior, replicator dynamics can be used to model how societal influence and financial costs can change opinions about resource extraction. Previous research on replicator dynamics has shown how evolving opinions on conservation can change how humans interact with their environment and therefore change population dynamics of the harvested species. However, social-ecological models often assume that human societies are homogeneous with no social structure. Building on previous work on social-ecological models, we develop a two-patch socio-ecological model with social hierarchy in order to study the interactions between spatial dynamics and social inequity. We found that fish movement between patches is a major driver of model dynamics, especially when the two patches exhibit different social equality and fishing practices. Further, we found that the societal influence between groups of harvesters was essential to ensuring stable fishery dynamics. Next, we developed a case study of two independently managed fisheries that were connected by fish movement where one human group fishes sustainably while another was over-harvests, resulting in a fishery collapse of both patches. We also found that because in this model, the influence of one human patch on another only communicates the amount of each catch and no fishing strategies were employed, increased social influence decreased the sustainability of the fishery. The findings of this study indicate the importance of including spatial components to socio- ecological models and highlights the importance of understanding species’ movements when making conservation decisions. Further, we demonstrate how incorporating fishing methods from outside sources can result in higher stability of the harvested population, demonstrating the need for effective communication across management regimes. 

   more » « less
4. Shapeshifting attachment: Exploring multi-dimensional people–place bonds in place-based citizen science

   https://doi.org/https://doi.org/10.1002/pan3.10174  

   Haywood, Benjamin; Parrish, Julia; He, Yurong (February 2021, People and nature)

   null (Ed.)

   Research on citizen science programmes has highlighted that they can foster science content and knowledge gain, enhance pro-environmental behaviour and cultivate civic action among participants. Especially in the case of place-based citizen science, which requires hands-on repeated activity in an out-of-door setting through a scientific lens, evidence suggests that some of these outcomes may be linked to the unique people–place relationships and interactions afforded by such programmes. Even still, studies that empirically examine the influence of place on citizen science participant and programme outcomes are scant. This is due, in part, to the methodological challenges involved in interrogating complex aspects of a person's sense of place—aspects like place attachment—the emotional bonds between people and place. Here, an adapted three-dimensional model of place attachment is proposed as a theoretical framework from which place-based citizen science experiences and outcomes might be empirically examined in depth. The model, which posits personal, social and natural environment dimensions of place attachment is contextualized with research findings from the US-based Coastal Observation and Seabird Survey Team (COASST) citizen science programme. Data from COASST suggest that participants do exhibit place attachment in all three dimensions of attachment, categorized within seven unique constructs, although questions remain regarding the unique intensity, make-up (shape) and scale (spatial, social and nature-science) of individual-level attachment along the three central dimensions. Critically, more research is needed to investigate whether the unique place attachment ‘profile’ of participants is a function of personal, social or programmatic variables pre- and post-programme participation. To encourage further scholarship on potential links between the experiences, exposures and programme components of place-based citizen science and the place attachment profiles of participants, this paper includes a brief review of the research opportunities presented by the adapted three-dimensional place attachment model discussed. Advancing this line of inquiry is an important component of broader efforts to understand how sense of place is altered via place-based citizen science and whether or not that is linked to specific programme outputs or participant outcomes in science knowledge, ecological understanding and civic engagement. 

   more » « less
5. Urbanization as a facilitator of gene flow in a human health pest

   https://doi.org/10.1111/mec.14783  

   Miles, Lindsay S.; Johnson, J. Chadwick; Dyer, Rodney J.; Verrelli, Brian C. (July 2018, Molecular Ecology)

   Abstract Urban fragmentation can reduce gene flow that isolates populations, reduces genetic diversity and increases population differentiation, all of which have negative conservation implications. Alternatively, gene flow may actually be increased among urban areas consistent with an urban facilitation model. In fact, urban adapter pests are able to thrive in the urban environment and may be experiencing human‐mediated transport. Here, we used social network theory with a population genetic approach to investigate the impact of urbanization on genetic connectivity in the Western black widow spider, as an urban pest model of human health concern. We collected genomewide single nucleotide polymorphism variation from mitochondrial and nuclear double‐digest RAD (ddRAD) sequence data sets from 210 individuals sampled from 11 urban and 10 nonurban locales across its distribution of the Western United States. From urban and nonurban contrasts of population, phylogenetic, and network analyses, urban locales have higher within‐population genetic diversity, lower between‐population genetic differentiation and higher estimates of genetic connectivity. Social network analyses show that urban locales not only have more connections, but can act as hubs that drive connectivity among nonurban locales, which show signatures of historical isolation. These results are consistent with an urban facilitation model of gene flow and demonstrate the importance of sampling multiple cities and markers to identify the role that urbanization has had on larger spatial scales. As the urban landscape continues to grow, this approach will help determine what factors influence the spread and adaptation of pests, like the venomous black widow spider, in building policies for human and biodiversity health. 

   more » « less