-
Deconvolving Complex Neuronal Networks into Interpretable Task-Specific Connectomes
Authors:
Yifan Wang,
Vikram Ravindra,
Ananth Grama
Abstract:
Task-specific functional MRI (fMRI) images provide excellent modalities for studying the neuronal basis of cognitive processes. We use fMRI data to formulate and solve the problem of deconvolving task-specific aggregate neuronal networks into a set of basic building blocks called canonical networks, to use these networks for functional characterization, and to characterize the physiological basis…
▽ More
Task-specific functional MRI (fMRI) images provide excellent modalities for studying the neuronal basis of cognitive processes. We use fMRI data to formulate and solve the problem of deconvolving task-specific aggregate neuronal networks into a set of basic building blocks called canonical networks, to use these networks for functional characterization, and to characterize the physiological basis of these responses by mapping them to regions of the brain. Our results show excellent task-specificity of canonical networks, i.e., the expression of a small number of canonical networks can be used to accurately predict tasks; generalizability across cohorts, i.e., canonical networks are conserved across diverse populations, studies, and acquisition protocols; and that canonical networks have strong anatomical and physiological basis. From a methods perspective, the problem of identifying these canonical networks poses challenges rooted in the high dimensionality, small sample size, acquisition variability, and noise. Our deconvolution technique is based on non-negative matrix factorization (NMF) that identifies canonical networks as factors of a suitably constructed matrix. We demonstrate that our method scales to large datasets, yields stable and accurate factors, and is robust to noise.
△ Less
Submitted 3 July, 2024; v1 submitted 28 June, 2024;
originally announced July 2024.
-
Agile Satellite Planning for Multi-Payload Observations for Earth Science
Authors:
Rich Levinson,
Sreeja Nag,
Vinay Ravindra
Abstract:
We present planning challenges, methods and preliminary results for a new model-based paradigm for earth observing systems in adaptive remote sensing. Our heuristically guided constraint optimization planner produces coordinated plans for multiple satellites, each with multiple instruments (payloads). The satellites are agile, meaning they can quickly maneuver to change viewing angles in response…
▽ More
We present planning challenges, methods and preliminary results for a new model-based paradigm for earth observing systems in adaptive remote sensing. Our heuristically guided constraint optimization planner produces coordinated plans for multiple satellites, each with multiple instruments (payloads). The satellites are agile, meaning they can quickly maneuver to change viewing angles in response to rapidly changing phenomena. The planner operates in a closed-loop context, updating the plan as it receives regular sensor data and updated predictions. We describe the planner's search space and search procedure, and present preliminary experiment results. Contributions include initial identification of the planner's search space, constraints, heuristics, and performance metrics applied to a soil moisture monitoring scenario using spaceborne radars.
△ Less
Submitted 13 November, 2021;
originally announced November 2021.
-
Attitude Trajectory Optimization for Agile Satellites in Autonomous Remote Sensing Constellation
Authors:
Emmanuel Sin,
Sreeja Nag,
Vinay Ravindra,
Alan Li,
Murat Arcak
Abstract:
Agile attitude maneuvering maximizes the utility of remote sensing satellite constellations. By taking into account a satellite's physical properties and its actuator specifications, we may leverage the full performance potential of the attitude control system to conduct agile remote sensing beyond conventional slew-and-stabilize maneuvers. Employing a constellation of agile satellites, coordinate…
▽ More
Agile attitude maneuvering maximizes the utility of remote sensing satellite constellations. By taking into account a satellite's physical properties and its actuator specifications, we may leverage the full performance potential of the attitude control system to conduct agile remote sensing beyond conventional slew-and-stabilize maneuvers. Employing a constellation of agile satellites, coordinated by an autonomous and responsive scheduler, can significantly increase overall response rate, revisit time and global coverage for the mission. In this paper, we use recent advances in sequential convex programming based trajectory optimization to enable rapid-target acquisition, pointing and tracking capabilities for a scheduler-based constellation. We present two problem formulations. The Minimum-Time Slew Optimal Control Problem determines the minimum time, required energy, and optimal trajectory to slew between any two orientations given nonlinear quaternion kinematics, gyrostat and actuator dynamics, and state/input constraints. By gridding the space of 3D rotations and efficiently solving this problem on the grid, we produce lookup tables or parametric fits off-line that can then be used on-line by a scheduler to compute accurate estimates of minimum-time and maneuver energy for real-time constellation scheduling. The Minimum-Effort Multi-Target Pointing Optimal Control Problem is used on-line by each satellite to produce continuous attitude-state and control-input trajectories that realize a given schedule while minimizing attitude error and control effort. The optimal trajectory may then be achieved by a low-level tracking controller. We demonstrate our approach with an example of a reference satellite in Sun-synchronous orbit passing over globally-distributed, Earth-observation targets.
△ Less
Submitted 15 February, 2021;
originally announced February 2021.
-
Planning a Reference Constellation for Radiometric Cross-Calibration of Commercial Earth Observing Sensors
Authors:
Sreeja Nag,
Philip Dabney,
Vinay Ravindra,
Cody Anderson
Abstract:
The Earth Observation planning community has access to tools that can propagate orbits and compute coverage of Earth observing imagers with customizable shapes and orientation, model the expected Earth Reflectance at various bands, epochs and directions, generate simplified instrument performance metrics for imagers and radars, and schedule single and multiple spacecraft payload operations. We are…
▽ More
The Earth Observation planning community has access to tools that can propagate orbits and compute coverage of Earth observing imagers with customizable shapes and orientation, model the expected Earth Reflectance at various bands, epochs and directions, generate simplified instrument performance metrics for imagers and radars, and schedule single and multiple spacecraft payload operations. We are working toward integrating existing tools to design a planner that allows commercial small spacecraft to assess the opportunities for cross-calibration of their sensors against current satellite to be calibrated, specifications of the reference instruments, sensor stability, allowable latency between calibration measurements, differences in viewing and solar geometry between calibration measurements, etc. The planner would output cross-calibration opportunities for every reference target pair as a function of flexible user-defined parameters. We use a preliminary version of this planner to inform the design of a constellation of transfer radiometers that can serve as stable, radiometric references for commercial sensors to cross-calibrate with. We propose such a constellation for either vicarious cross-calibration using pre-selected sites, or top of the atmosphere (TOA) cross-calibration globally. Results from the calibration planner applied to a subset of informed architecture designs show that a 4 sat constellation provides multiple calibration opportunities within half a day planning horizon, for Cubesat sensors deployed into a typical rideshare orbits. While such opportunities are available for cross calibration image pairs within 5 deg of solar or view directions, and with-in an hour (for TOA) and less than a day (vicariously), the planner allows us to identify many more by relaxing user-defined restrictions.
△ Less
Submitted 19 October, 2020;
originally announced October 2020.
-
Autonomous Scheduling of Agile Spacecraft Constellations with Delay Tolerant Networking for Reactive Imaging
Authors:
Sreeja Nag,
Alan S. Li,
Vinay Ravindra,
Marc Sanchez Net,
Kar-Ming Cheung,
Rod Lammers,
Brian Bledsoe
Abstract:
Small spacecraft now have precise attitude control systems available commercially, allowing them to slew in 3 degrees of freedom, and capture images within short notice. When combined with appropriate software, this agility can significantly increase response rate, revisit time and coverage. In prior work, we have demonstrated an algorithmic framework that combines orbital mechanics, attitude cont…
▽ More
Small spacecraft now have precise attitude control systems available commercially, allowing them to slew in 3 degrees of freedom, and capture images within short notice. When combined with appropriate software, this agility can significantly increase response rate, revisit time and coverage. In prior work, we have demonstrated an algorithmic framework that combines orbital mechanics, attitude control and scheduling optimization to plan the time-varying, full-body orientation of agile, small spacecraft in a constellation. The proposed schedule optimization would run at the ground station autonomously, and the resultant schedules uplinked to the spacecraft for execution. The algorithm is generalizable over small steerable spacecraft, control capability, sensor specs, imaging requirements, and regions of interest. In this article, we modify the algorithm to run onboard small spacecraft, such that the constellation can make time-sensitive decisions to slew and capture images autonomously, without ground control. We have developed a communication module based on Delay/Disruption Tolerant Networking (DTN) for onboard data management and routing among the satellites, which will work in conjunction with the other modules to optimize the schedule of agile communication and steering. We then apply this preliminary framework on representative constellations to simulate targeted measurements of episodic precipitation events and subsequent urban floods. The command and control efficiency of our agile algorithm is compared to non-agile (11.3x improvement) and non-DTN (21% improvement) constellations.
△ Less
Submitted 19 October, 2020;
originally announced October 2020.
-
De-anonymization Attacks on Neuroimaging Datasets
Authors:
Vikram Ravindra,
Ananth Grama
Abstract:
Advances in imaging technologies, combined with inexpensive storage, have led to an explosion in the volume of publicly available neuroimaging datasets. Effective analyses of these images hold the potential for uncovering mechanisms that govern functioning of the human brain, and understanding various neurological diseases and disorders. The potential significance of these studies notwithstanding,…
▽ More
Advances in imaging technologies, combined with inexpensive storage, have led to an explosion in the volume of publicly available neuroimaging datasets. Effective analyses of these images hold the potential for uncovering mechanisms that govern functioning of the human brain, and understanding various neurological diseases and disorders. The potential significance of these studies notwithstanding, a growing concern relates to the protection of privacy and confidentiality of subjects who participate in these studies. In this paper, we present a de-anonymization attack rooted in the innate uniqueness of the structure and function of the human brain. We show that the attack reveals not only the identity of an individual, but also the task they are performing, and their efficacy in performing the tasks. Our attack relies on novel matrix analyses techniques that are used to extract discriminating features in neuroimages. These features correspond to individual-specific signatures that can be matched across datasets to yield highly accurate identification. We present data preprocessing, signature extraction, and matching techniques that are computationally inexpensive, and can scale to large datasets. We discuss implications of the attack and challenges associated with defending against such attacks.
△ Less
Submitted 8 August, 2019;
originally announced August 2019.
