169 results about "Physics based" patented technology

A markerless method is described for tracking the motion of subjects in a three dimensional environment using a model based on linked kinematic chains. The invention is suitable for tracking robotic, animal or human subjects in real-time using a single computer with inexpensive video equipment, and does not require the use of markers or specialized clothing. A simple model of rigid linked segments is constructed of the subject and tracked using three dimensional volumetric data collected by a multiple camera video imaging system. A physics based method is then used to compute forces to align the model with subsequent volumetric data sets in real-time. The method is able to handle occlusion of segments and accommodates joint limits, velocity constraints, and collision constraints and provides for error recovery. The method further provides for elimination of singularities in Jacobian based calculations, which has been problematic in alternative methods.

Disclosed are methods and systems for a virtual reality simulation and display of limb movement that facilitate the development and fitting of prosthetic control of a paralyzed or artificial limb. The user generates command signals that are then processed by the control system. The output of the control system drives a physics-based simulation of the limb that simulates the limb to be controlled. The computed movements of the model limb are displayed to the user as a 3D animation from the perspective of the user so as to give the impression that the user is watching the actual movements of his/her own limb. The user learns to adjust his/her command signals to perform tasks successfully with the virtual limb. Alternatively or additionally, the errors produced by the virtual limb and/or the responses of the user during the training process can provide information for adapting the properties of the control system itself.

A method and system for performing continuous (real-time) physics based prognostics analysis as a function of actual engine usage and changing operating environment. A rule-based mission profile analysis is conducted to determine the mission variability which yields variability in the type of thermal-mechanical loads that an engine is subjected to during use. This is followed by combustor modeling to predict combustion liner temperatures and combustion nozzle plane temperature distributions as a function of engine usage which is followed by off-design engine modeling to determine the pitch-line temperatures in hot gas path components and thermodynamic modeling to compute the component temperature profiles of the components for different stages of the turbine. This is automatically followed by finite element(FE) based non-linear stress-strain analysis using an real-time FE solver and physics based damage accumulation, life consumption and residual life prediction analyses using microstructural modeling based damage and fracture analysis techniques.

A system, method and computer program product is provided for interactive user navigation in a real-time 3D simulation. An assembly builder permits a user to build customized physics-based assemblies for user navigation in a variety of virtual environments. These assemblies are stored in a library and are then accessed by a navigation run-time module that runs in conjunction with, or as a part of, a visual run-time application. The navigation run-time module receives high-level user goal requests via a simple and intuitive user interface, converts them into a series of tasks, and then selects the appropriate assembly or assemblies to perform each task. As a result, complex navigation may be achieved. Once selected, an assembly provides a physics-based eye-point model for user navigation. Collisions between the assembly and objects in the simulation are resolved using a real-time physics engine, thus ensuring smooth, cinematic-style eye-point modeling in addition to real-time control.

The present disclosure relates to the use of hybrid predictive models to predict one or more of performance, availability, or degradation of a power plant or a component of the power plant. The hybrid predictive model comprises at least two model components, one based on a physics-based modeling approach and one based on an observational or data-based modeling approach. The hybrid predictive model may self-tune or self-correct as operational performance varies over time.

Methods for three-dimensionally characterizing a reservoir while drilling a high angle or horizontal wellbore through the reservoir are disclosed. An initial reservoir model for the reservoir is selected and a section is extracted for a planned trajectory of the wellbore. A secondary model is generated by performing secondary modeling for at least part of the planned trajectory. An area of interest is identified within the secondary model where statistical uncertainty is high. Possible causes of the statistical uncertainty are identified for the area of interest within the secondary model that are not present or accounted for in the initial reservoir model. A set of parameters for the area of interest are defined at that are based on the possible causes of statistical uncertainty. The area of interest is logged with at least one logging while drilling LWD tool. Sensitivities of the LWD tool response to the subset of parameters are evaluated by performing at least one tertiary model for a range of the subset of parameters. The most sensitive parameters from the subset of parameters and corresponding measurements are identified. One or more real-time LWD measurements to be used for proactive well placement along the planned trajectory are identified and are based on the most sensitive parameters. The initial reservoir model is updated while drilling with information from the tertiary model. The model update is based on physics-based modeling or on inversion and on running multiple models and selection of a best candidate model based on correlations between the tool measurements and modeled results for each geologic model.

A distributed simulation system is composed of simulator stations linked over a network that each renders real-time video imagery for its user from scene data stored in its data storage. The simulation stations are each connected with a physics farm that manages the virtual objects in the shared virtual environment based on their physical attribute data using physics engines, including an engine at each simulation station. The physics engines of the physics farm are assigned virtual objects so as to reduce the effects of latency, to ensure fair fight requirements of the system, and, where the simulation is of a vehicle, to accurately model the ownship of the user at the station. A synchronizer system is also provided that allows for action of simulated entities relying on localized closed loop controls to cause the entities to meet specific goal points at specified system time points.

A well monitoring system particularly useful in detecting kicks in the well includes a well, a well system, and a computing apparatus. The well defines a wellbore and the well system includes at least one sensor measuring at least one well condition. The computing apparatus hosts a well monitoring software component that performs a method to detect a kick in a well. The method includes: storing a set of real-time data from a measurement of a well condition by the sensor, the measurements being correlative to an unplanned fluid influx into the well; modeling the operation of the well with a physics-based, state space model of the well system to obtain an estimate of the well condition; and applying the real-time data set and the estimate to a probabilistic estimator to yield a probability of an occurrence of a kick and a confidence measure for the probability.

A computer simulation enhanced exercise device is provided which engages the user by directly relating the users exercise motion in real time to a visual simulation or interactive game. The exercise device may comprise any variety of machines including, stationary bikes, rowing machines, treadmills, stepper, elliptical gliders or under desk exercise. These exercise devices are configured with sensors to measure physical movements as the user exercises and are coupled to computer hardware with modeling and virtualization software to create the system. These sensor measurements are then sent to a computer for use in the physics based modeling and real-time visual simulation. The computer simulation enhanced exercise device is further provided with features including manual and automatic adjustment of resistance levels, visualization of accurate caloric burn rates and correlation to everyday food items, and network connectivity providing for multiplayer network simulations and directed advertising.

A method for modeling changes in the state of charge vs. open circuit voltage (SOC-OCV) curve for a lithium-ion battery cell as it ages. During battery pack charging, voltage and current data are gathered for a battery cell. A set of state equations are used to determine the stoichiometry and state of charge of the cathode half-cell based on the charging current profile over time. The voltage and current data, along with the stoichiometry and state of charge of the cathode half-cell, are then used to estimate maximum and minimum solid concentration values at the anode, using an error function parameter regression/optimization. With stoichiometric conditions at both the cathode and anode calculated, the cell's capacity and a new SOC-OCV curve can be determined.

A process and a system for assisting a clinician in identification and evaluation of perfusion characteristics of an organ are disclosed. The process and system may be used to extend clinical diagnosis capabilities by augmenting the ability to analyze perfusion images acquired from a radiology imaging modality through the use of the quantification technology as described. Classification of perfusion characteristics of a region of interest is facilitated through the use of a quantitative spatial perfusion characteristics map. The map is generated by a computing device which applies at least one physics-based image processing technique. This map is displayed on a clinician device for review by the clinician to assist in the spatial identification of perfusion abnormalities.

A system for using machine-learning to create a model for performing integrated circuit layout extraction is disclosed. The system of the present invention has two main phases: model creation and model application. The model creation phase comprises creating one or more extraction models using machine-learning techniques. First, a complex extraction problem is decomposed into smaller simpler extraction problems. Then, each smaller extraction problem is then analyzed to identify a set of physical parameters that fully define the smaller extraction problem. Next, models are created using machine learning techniques for all of the smaller simpler extraction problems. The machine learning is performed by first creating training data sets composed of the identified parameters from typical examples of the smaller extraction problem and the answers to those example extraction problems as solved using a highly accurate physics-based field solver. The training sets are then used to train the models. In one embodiment, neural networks are used to model the extraction problems. Bayesian inference is employed by one embodiment in order to train the neural network models. Bayesian inference may be implemented with normal Monte Carlo techniques or Hybrid Monte Carlo techniques. After the creation of a set of models for each of the smaller simpler extraction problems, the machine-learning based models may be used for extraction.

A battery state estimator which combines an electrochemical solid-state concentration model with an empirical equivalent-circuit model. The battery state estimator uses a reduced-order physics-based electrochemical solid-state concentration model to calculate open circuit voltage of a battery cell, and uses the calculated open circuit voltage in an equivalent-circuit model to evaluate a diffusion voltage drop and other battery cell parameters. The battery state estimator is implemented in an online estimator framework using an extended Kalman filter, with a measured terminal voltage and measured current as inputs. A state of charge of the battery cell can be determined from the calculated open circuit voltage, and the state of charge along with the other parameters from the equivalent-circuit model are used to evaluate battery pack performance and to control battery pack charging and discharging.

A system for using machine-learning to create a model for performing integrated circuit layout extraction is disclosed. The system of the present invention has two main phases: model creation and model application. The model creation phase comprises creating one or more extraction models using machine-learning techniques. First, a complex extraction problem is decomposed into smaller simpler extraction problems. Then, each smaller extraction problem is then analyzed to identify a set of physical parameters that fully define the smaller extraction problem. Next, models are created using machine learning techniques for all of the smaller simpler extraction problems. The machine learning is performed by first creating training data sets composed of the identified parameters from typical examples of the smaller extraction problem and the answers to those example extraction problems as solved using a highly accurate physics-based field solver. Next, the system trains a set of neural networks using the training sets. In one embodiment, Bayesian inference is used to train the neural networks that are used to model the extraction. After the creation the neural network based models for each of the smaller simpler extraction problems, the neural network based models may be used for extraction.

A basic component of Computer-Aided Detection systems for digital mammography comprises generating candidate mass locations suitable for further analysis. A component is described that relies on filtering either the background image or the complementary foreground mammographic detail by a purely signal processing method on the one hand or a processing method based on a physical model on the other hand. The different steps of the signal processing approach consist of band-pass filtering the image by one or more band pass filters, multidimensional clustering, iso-contouring of the distance to centroid of the one or more filtered values, and finally candidate generation and segmentation by contour processing. The physics-based approach also filters the image to retrieve a fat-corrected image to model the background of the breast, and the resulting image is subjected to a blob detection filter to model the intensity bumps on the foreground component of the breast that are associated with mass candidates.

The disclosure provides an approach for learning to schedule control fragments for physics-based virtual character simulations and physical robot control. Given precomputed tracking controllers, a simulation application segments the controllers into control fragments and learns a scheduler that selects control fragments at runtime to accomplish a task. In one embodiment, each scheduler may be modeled with a Q-network that maps a high-level representation of the state of the simulation to a control fragment for execution. In such a case, the deep Q-learning algorithm applied to learn the Q-network schedulers may be adapted to use a reward function that prefers the original controller sequence and an exploration strategy that gives more chance to in-sequence control fragments than to out-of-sequence control fragments. Such a modified Q-learning algorithm learns schedulers that are capable of following the original controller sequence most of the time while selecting out-of-sequence control fragments when necessary.

An embodiment of a method for lifespan modeling for a turbine engine component includes determining a design-phase model of the lifespan of an turbine engine component; fusing the design-phase model with sensor data collected during operation of the turbine engine component to produce an updated model of the lifespan of the turbine engine component; and fusing the updated model with data collected during an inspection of the turbine engine component to produce an overall model of the lifespan of the turbine engine component. Systems for lifespan modeling for a turbine engine component are also provided.

A system for using machine-learning to create a model for performing integrated circuit layout extraction is disclosed. The system of the present invention has two main phases: model creation and model application. The model creation phase comprises creating one or more extraction models using machine-learning techniques. First, a complex extraction problem is decomposed into smaller simpler extraction problems. Then, each smaller extraction problem is then analyzed to identify a set of physical parameters that fully define the smaller extraction problem. Next, models are created using machine learning techniques for all of the smaller simpler extraction problems. The machine learning is performed by first creating training data sets composed of the identified parameters from typical examples of the smaller extraction problem and the answers to those example extraction problems as solved using a highly accurate physics-based field solver. The training sets are then used to train the models. In one embodiment, neural networks are used to model the extraction problems. To train the neural network models. Bayesian inference is used in one embodiment. Bayesian inference may be implemented with normal Monte Carlo techniques or Hybrid Monte Carlo techniques. After the creation of a set of models for each of the smaller simpler extraction problems, the machine-learning based models may be used for extraction.

A system for using machine learning based upon Bayesian inference using a hybrid monte carlo method to create a model for performing integrated circuit layout extraction is disclosed. The system of the present invention has two main phases: model creation and model application. The model creation phase comprises creating one or more extraction models using machine-learning techniques. First, a complex extraction problem is decomposed into smaller simpler extraction problems. Then, each smaller extraction problem is then analyzed to identify a set of physical parameters that fully define the smaller extraction problem. Next, complex mathematical models are created using machine learning techniques for all of the smaller simpler extraction problems. The machine learning is performed by first creating training data sets composed of the identified parameters from typical examples of the smaller extraction problem and the answers to those example extraction problems as solved using a highly accurate physics-based field solver. Next, the system uses Bayesian inference implemented with a Monte Carlo method to train a set of neural networks for extraction problems. After the creation of a set of models for each of the smaller simpler extraction problems, the machine-learning based models may be used for extraction.