93887 research outputs found
Sort by
Demonstration of a Digital Twin framework for a two-actuator hydraulic application
Get PDFThis project presents a digital twin framework for controlling a hydraulic crane using AI, mixed reality, and real-time actuation. Leveraging a Jetson Nano and Xbox Kinect for perception and a Raspberry Pi with DRV103 for control, the system enables adaptive motion planning and obstacle-aware navigation. A Unity-based interface integrated with HoloLens2 allows operators to visualize and manipulate the crane through FABRIK inverse kinematics. Real-time environmental mapping from Kinect enables obstacle detection to update the motion path dynamically. Early experiments confirm the ability to compute joint angles in real-time, detect environmental objects, and demonstrate closed-loop control in a virtual-physical hybrid system
Bilateral Symmetric 3D Reconstructor (BiaSed): A 3-Dimensional Object Re- construction Model Demonstrating Human-Like Shape Constancy
Get PDFInfluence of Task on the Geometry of a Perceptual Space
Get PDFPerceptual spaces are representations of a sensory or cognitive domain in which the domain’s elements correspond to points, and distances between these points are perceptual differences. Proximity relationships within a perceptual space can support a variety of functions, including discrimination, grouping, learning, and generalization. These diverse functions may use the features of the domain in different ways, resulting in task-dependent influences on the geometry of the space.
To identify and characterize these influences, we focused on a domain of visual textures. These textures varied across many dimensions, including mean luminance and low- and high-order spatial correlations, which formed the axes of the space. Previous work characterized the geometry of this space with a threshold texture segmentation task: an approximately Euclidean distance that corresponded to the informativeness of the image statistics in natural images. However, when subjects are asked to make suprathreshold similarity judgments, the geometry of the space changes in two ways: greater weight was given to the higher-order local features (linear transformation), and axes became curved (nonlinear transformation).
Here we report that the geometry undergoes further transformations with specific tasks: judging similarity based on brightness, judging similarity based on visual working memory, and grouping. These changes are consistent across subjects (N= 6) and primarily consist of linear transformations. This reformatting likely represents top-down influences on the gains of neural populations
Using Neural Networks to Better Understand Static and Dynamic Components of Facial Expression Recognition
Get PDFFacial expressions are crucial social information for human communication. In the real world, facial expressions are dynamic; however, much of existing research in facial expressions relies on static stimuli. This static approach may limit the ecological validity of our understanding of the emotional information on faces. Our study aims to investigate the dynamic and static information contained in different emotional categories of facial expressions (e.g., happy, sad). Four convolutional neural networks are introduced as models to be trained with a large-scale dynamic facial expression dataset (short videoclips of 16 frames of 7 different emotional categories) in four different ways: ordered frames (of a single videoclip,16 at a time), shuffled frames, ordered global temporal change (15 optical flows or dynamic changes between adjacent frames, of a single videoclip in correct global sequence), and shuffled local temporal change. We compare model performance across these different training regimes in their ability to accurately identify the different emotional categories. Our results show that local dynamic information contributes to the recognition of sad and angry, and to a lesser extent fear, expressions, and global dynamic information contributes to the recognition of surprise and happy, and to a lesser extent disgust and fear, expressions. As expected, static structural information contributed to the recognition of sad, neutral, happy, surprise, and fear expressions. By highlighting the differential contributions of dynamic and static information, this study emphasizes the need for more ecologically valid approaches in the study of facial expression recognition
Capturing the Representational Dynamics of Face Perception in Deep Recurrent Neural Networks
Get PDFWe investigate the representational dynamics of recurrent convolutional neural networks (RCNNs) in order to understand the time-course of visual recognition. We explore a family of models with bottom-up and lateral connections that were optimized for face-identification and object-recognition tasks. Using representational similarity analysis (RSA), we observed that only models that were trained for face identification showed a late-emerging prominent distinction of identities as seen in the monkey face patch AM. Early model responses strongly separated the objects from the faces. These findings suggest that the dynamics of face recognition that emerges in a hierarchical recurrent neural network prioritizes category-level recognition at early stages (face detection), triggering later category-specific computations that enable individual-level recognition (face recognition) as observed in neurophysiological findings. Our results also show that models that were trained simultaneously on both face identification and object recognition were more likely to show the signature of mirror symmetric viewpoint tuning in their intermediate representations as has been reported for monkey face patch AL. We also examined the tuning properties of individual units in the last layer of our network across timesteps. After embedding the face/non-face images in a representational space, for each unit the tuning was determined as the direction in which the unit responses increased. With increasing steps, the units showed an emerging identity discrimination tuning that was recently observed in primate face patches. Taken together, these results give us a candidate mechanistic account of primate face perception, consistent with evidence on individual unit tuning and population geometry
Reduced Order Approach For Peening Stress Field Variability
No full textA common goal in shot peening research is to connect operational parameters to resultant residual stress fields, providing a means to control and optimize the effectiveness of surface treatment. In practice, experimental measurements of residual stresses are often averaged values over regions that are large in comparison to an impact dimple. In fact, the stochastic nature of impact locations leads to residual stress fields that are distributed. Finite element peening simulations confirm this observation. The goal of this report is to connect operational parameters to localized fluctuations in residual stress through probabilistic reasoning (Figure 1). In particular, the development of a Poisson process model to predict variability in residual stresses over measurement regions comprising multiple impacts, as well as predicting asymptotic variability in residual stress at sub-impact measurement length scales
Adaptation Of Shot Peen Parameters For Gear Geometry
No full textShot peening is a well established process for the surface enhancement of gearing. Gearing is a primary example of a high cycle fatigue application that can benefit from residual stress enhancement. Designing shot peening parameters to specific gear geometry based on material, heat treatment, and surface finish is a more precise way to achieve better performance outcomes. One of the best tools to assist in the optimal shot peening is x-ray diffraction (XRD) and its ability to measure small differences that can result in significant performance outcomes. XRD residual stress measurements are a direct measurement of elastic strain. The diffraction peak width is an indication of plastic strain and is also proportional to the hardness of steels. Together, the elastic and plastic strain information provides tremendous insight into the condition of a shot peened gear.
This paper will review a wide variety of gear geometry and their various heat treatments and surface finishes. The discussion will encompass how the shot peen parameters can be modified based on the physical characteristics of the gear and also the failure mode. X-ray diffraction results will be provided to supplement the discussion